Infectious Disease Epidemiology

Infectious disease epidemiology (which includes the epidemiology of viruses) is the study of the complex relationships among hosts and infectious agents.

From: Viruses , 2017

Pathogen Epidemiology

W.P. Hanage , in Encyclopedia of Evolutionary Biology, 2016

Conclusion

Infectious disease epidemiology is a practical science, concerned with minimizing the impact of pathogens on public health. As both pathogens and their hosts have evolved, evolutionary biology is relevant to understanding the nature of their interactions for fitness, and also in resolving the history of pathogen transmission. Mathematical models can explore the consequences of different selective scenarios, and molecular data can define strains and the genetic variation that is the raw material on which natural selection acts. Recent advances, especially in the rapid determination of sequence data, are bringing evolutionary biology ever closer to the clinic.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780128000496002286

Virus Transmission and Epidemiology

Susan Payne , in Viruses, 2017

Infectious disease epidemiology (which includes the epidemiology of viruses) is the study of the complex relationships among hosts and infectious agents. Epidemiologists are interested in virus spread or transmission, with or without disease. Viral epidemiologists try to predict the potential for development of epidemics, and a very important part of their job is to define the kinds of interventions that could contain a virus outbreak. Veterinarians are often concerned with threats to food animals (how a disease of food animals might be spread, or be introduced into a disease-free area). In order to model virus transmission, epidemiologists must try to account for a variety of factors involving both host and virus. Factors that can impact virus transmission and spread include:

•

Prevalence of the agent within the population.

•

Mode or method of transmission of the agent.

•

Duration of the infection and the window of transmissibility.

•

Numbers of susceptible and nonsusceptible individuals in the population.

•

Population density.

•

Patterns of travel or associations (for example, schoolchildren and their families form interconnected networks).

•

Living conditions.

•

Climate and/or season.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780128031094000052

Social Dimensions of Infectious Diseases

Johannes Sommerfeld , in International Encyclopedia of Public Health (Second Edition), 2017

Risk and Vulnerability

Traditional infectious disease epidemiology has focused, to a large extent, on measurable biological and behavioral risk factors at the level of the individual. Within the perspective of 'risk factor epidemiology,' epidemiologists focus their analytical work on proximate, individual-level risk factors. Social scientists, on the other hand, have a particular interest in elucidating the social context of risk and vulnerability to infectious diseases.

Proximate behavioral risk factors that shape infection, occurrence, and severity of the disease include individual and group behavior related, for example, to hygiene and sanitation, sexual behavior, food consumption, or movement (including travel, migration, and displacement). Social scientists emphasize the 'cultural logic' underpinning individual behavior and argue that behavioral patterns, and, consequently, exposure to, and distribution of infectious disease risk, are the expression of larger-scale forces such as poverty, social inequalities, armed conflict, and other forms of social, economic, and political forces. To social scientists, basic notions such as 'risk group,' 'patient compliance,' and 'community' do not adequately grasp the complex social, cultural, and economic reality of populations. Consequently, in the social science literature, the social construct of vulnerability has widely replaced the epidemiological construct of 'risk,' providing a theoretical background for practical interventions that try to take into account how specific social contexts influence individual identity-constructions, while avoiding the risk of group discrimination.

Vulnerability to infectious disease results from several major overlapping factors, including socioeconomic, biological, and environmental factors. Macro-level social processes such as globalization and trade liberalization, unplanned rapid urbanization, widespread poverty, and inequalities lead to vulnerability of population subgroups.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780128036785004161

Organization of Public Health Systems

Theodore H. Tulchinsky MD, MPH , Elena A. Varavikova MD, MPH, PhD , in The New Public Health (Third Edition), 2014

•

Registration and vital statistics

•

Epidemiology of infectious diseases

•

Maintaining documentation and reports as required by the government, e.g., fiscal records, reportable diseases, inspection and laboratory reports

•

Health education and health promotion

•

Environmental protection and sanitation

•

Control of communicable diseases, sexually transmitted infections, human immunodeficiency virus, tuberculosis

•

Preventive prenatal, infant, and toddler care

•

Coordination and cooperation with Departments of Education, Social Welfare, Agriculture, Environmental Protection, Urban Planning, and others

•

Allocation of resources

•

Planning and management of services

•

Licensing and supervision of health facilities

•

Hospitals and home care

•

Care of disabled

•

Rehabilitation and long-term care

•

Coordination of health services

•

Intersectoral cooperation

•

Mental health

•

Emergency and disaster preparedness

•

Social assistance

•

Nutrition, including licensing of food establishments

•

Community participation advocacy

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780124157668000100

Biological Agents

Ian Greaves FRCP, FCEM, FIMC, RCS(Ed), DTM&H, DMCC, DipMedEd, RAMC , Paul Hunt MBBS, DipIMC(RCSEd), MCEM, MRCSEd, DMCC, RAMC , in Responding to Terrorism, 2010

Detection of biological agents

The UK framework consists of local infectious disease epidemiology services staffed with consultants in communicable disease control (CCDC) and informed by local health services. Collation of locally collected data enables the recognition of any new disease trends. Regional laboratories maintained by the HPA provide diagnostic services and further epidemiological expertise on the clinical and organisational management of outbreaks. The HPA also maintains reference laboratories with expertise on specific microbes.

Deliberate biological releases may be declared or covert. Responses to declared releases will be coordinated centrally and extra resources supplied to the affected region or regions as required. Detection of a covert release will only be possible once the first cases of infection begin to arise. The appearance of diseases that rarely occur in nature may alert to the possibility of a covert deliberate release. For example, HPA guidelines state that deliberate release should be considered as a cause in the event of a single case of inhalational anthrax. Once the causative agent has been identified, all individuals exposed need to be traced, decontaminated and offered appropriate treatment. This response is coordinated via an outbreak control team consisting of the CCDC and appropriate subject matter experts.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780080450438000040

Omics Analyses in Molecular Epidemiologic Studies

Betsy Foxman , in Molecular Tools and Infectious Disease Epidemiology, 2012

7.7 Analyses of Microbiome Data

The potential contributions of the omics to infectious disease epidemiology are not limited to studies of single organisms. Next generation sequencing technologies like the 454 pyrosequencing (Roche, Branford, CT) and massively parallel sequencing by synthesis (Illumina, San Diego, CA) make it possible to characterize the membership and structure of microbial communities based on genetic sequence. Studies use the sequence of the ribosomal genes (designated rDNA for ribosomal DNA), which are present in all cellular organisms, to characterize the communities. All cells have ribosomes, and the genes are highly conserved so they can be used for taxonomy. Rapid sequencing technologies give an unprecedented opportunity to characterize the microbial communities living on and in the human body, how they vary over short and long periods, respond to treatment, and interact with pathogens. The Human Microbiome Project is an ongoing effort funded by the National Institutes of Health to provide these parameters. The goals are:

•

Determining whether individuals share a core human microbiome,

•

Understanding whether changes in the human microbiome can be correlated with changes in human health,

•

Developing the new technological and bioinformatic tools needed to support these goals, and

•

Addressing the ethical, legal and social implications raised by human microbiome research (http://nihroadmap.nih.gov/hmp/).

Taxonomy based on ribosomal gene sequence does not map perfectly to taxonomy based on other methods. Bacteria were originally placed into taxonomic groupings based on phenotype. Genetic analyses have clearly demonstrated that speciation based on phenotype has only a weak relationship with groupings based on genotype. Grouping bacteria based on the sequence of the 16S rRNA gene that codes for the ribosome, or only part of this gene, results in groups that correspond only at a very general level with taxonomic groupings. Although the distribution of the resulting groups can be informative, the relationship of a group to a species may be minimal at best. Therefore the resulting groups are referred to as operational taxonomic units (OTUs), that is, groupings used in lieu of species (Table 7.6). If sequences are classified based on genetic relationships using phylogenetics, the OTUs are referred to as phylotypes.

Table 7.6. Ecological Terms and Parameters Used in Genetic Analysis of Microbial Communities

Operational Taxonomic Unit (OTU) Classification that distributes individuals into groups based on some criteria
Phylotype Classification based on phylogenetic relationships
Community An assemblage of different species in the same place and time
Metacommunity The larger community that encompasses spatially separated communities
Population A group of individuals of the same species in the same place and time
Metapopulation The larger population that encompasses spatially separated populations
Structure A description of the OTUs in a community and their relative abundance
Richness The number of different OTUs in a community
Evenness A measure indicating the relative amounts of each OTU in a community
Diversity A measure indicating both the richness and evenness of the community

The microbiota (also known as microbial flora) living on and in the human body contains bacteria, archaea, eukaryotes, and viruses. A microbial community refers to the mix of organisms present at a time and place, such as the microbial community living in the human nose, which can include several different bacterial and viral species. This is in contrast to a population, which refers to a single species or OTU. Staphylococcus aureus can be isolated from the nose, armpit, pubis, vagina, and bowel. The S. aureus in the nose is a population; presuming that nasal S. aureus is swallowed and wiped on hands and thus interacts, the different populations of S. aureus on a human body are a metapopulation. Similarly, the microbial communities in the nose interact with communities at other body sites; the microbiota on a human are a metacommunity.

The structure of a microbial community is a measure of the OTUs present and their relative abundance. Ecologists ask questions regarding the resistance of a community to disruption and the resilience following disruption. For epidemiologists, this translates into asking questions such as the following. (1) What is the role of nasal microbiota in preventing colonization by a potential pathogen? (2) Does existing microbiota enhance or reduce risk of pathogen transmission? (3) If the nasal microbiota is disrupted by an invading pathogen (and additionally by treatment of that pathogen), does it return to its previous state? To summarize these changes, populations are characterized with respect to richness, evenness, and diversity.

Richness refers to the total number of OTUs present. Because the number of OTUs observed depends on the number of individuals sampled, to compare species richness in different samples, data are displayed in a rarefaction curve. Rarefaction curves plot the number of OTUs observed by the number sampled. The richness can then be compared. For pyrosequencing, the number sampled would be the number of sequences read. Claesson and associates 23 analyzed the intestinal microbiota in four fecal samples using pyrosequencing of two different regions of the 16S rRNA, V4 and V6. The V6 region is known to be more variable, which therefore results in more OTUs than a less variable region. The rarefaction curve shows that for all four individuals (A, B, C, and D) at the same number of reads (horizontal line), there are more phylotypes observed for the V6 than V4 region (Figure 7.7). As a validation, the authors randomly selected half the reads and recreated the rarefaction curve (inset). This analysis confirmed that the V6 region is more variable.

Figure 7.7. Rarefaction curves for pyrosequencing reads of the V4 and V6 regions of 16S rDNA from fecal samples from individuals A, B, C, and D. Phylotypes assigned using 97% (dotted lines) and 98% levels (solid lines, except for ALL-V4, which has single dots) of similarity. The inset shows curves for half the A-V6-1.0 reads and the three constituent parts of the C-V4-0.5 reads.

 Source: Reproduced, with permission, from Claesson et al. 23 (2009).

Evenness refers to the distribution of each OTU within the sample. High evenness suggests there are similar numbers in each OTU and low evenness suggests the reverse. In the study by Claesson and associates, 23 the evenness varied from 0.51 to 0.70, using the same region and level of similarity for classification into phylotypes. Sample B was most even and sample D the least. Diversity is a function of richness and evenness; a frequently used index of diversity is the Shannon index. Diversity was highest in sample A and lowest in Sample D.

These technologies and analyses have applications beyond characterizing human microbiota. They can be used to assess environmental exposures, such as microbial composition in air pollutants or water, or infectious content on fomites in clinical settings. They can be used to measure specific genes, such as those for antimicrobial resistance or virulence within microbial communities independent of species. Although it is hard to predict how an increased understanding of microbial community structure will change epidemiology and public health practice, it is a given that we are in for some surprises.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780123741332000071

Epidemiologic Investigation for Public Health, Biodefense, and Forensic Microbiology

STEPHEN A. MORSE , ALI S. KHAN , in Microbial Forensics, 2005

MOLECULAR STRAIN TYPING

The microbiology laboratory has made significant contributions to the epidemiology of infectious diseases. The repeated isolation of a specific microorganism from patients with a given disease or syndrome has helped to prove infectious etiologies. In addition, the isolation and identification of microorganisms from animals, vectors, and environmental sources has been invaluable in identifying reservoirs and verifying modes of transmission. In dealing with an infection, it is often necessary to identify the species of the infecting microorganism in order to prescribe effective therapy. Many of the techniques that have evolved for such purposes are both rapid and accurate but, in general, do not provide the kind of genetic discrimination necessary for addressing epidemiologic questions. The epidemiology of many infectious diseases is becoming more complex. Fortunately, typing methods for bacteria, fungi, protozoa, and viruses have evolved to meet this challenge. Historically, the typing methods that have been used in epidemiologic investigations fall into two broad categories: phenotypic methods and genotypic methods. Phenotypic methods are those methods that characterize the products of gene expression in order to differentiate strains. For example, the use of biochemical profiles to discriminate between genera and species of bacteria is used as a diagnostic method, but can also be used for biotyping. Other methods, such as phage typing, can be used to discriminate among groups within a bacterial species. Biotyping emerged as a useful tool for epidemiologic investigations in the 1960s and early 1970s, while phage typing of bacteria and serological typing of bacteria and viruses has been used for decades. Today, the majority of these tests are considered inadequate for epidemiologic purposes. First, they do not provide enough unrelated parameters to obtain a good reflection of genotype. For example, serotyping of Streptococcus pneumoniae discriminates among only a limited number of groups. In addition, some virus species, such as human cytomegalovirus and measles virus, cannot be divided into different types or subtypes by serology, because significant antigenic differences do not exist. Second, the expression of many genes is affected by spontaneous mutations, environmental conditions, and by developmental programs or reversible phenotypic changes, such as high-frequency phenotypic switching. Because of this, many of the properties measured by phenotypic methods have a tendency to vary, and for the most part they have been replaced by genotypic methods. The one major exception is multilocus enzyme electrophoresis (MLEE), 19, 20 which is a robust phenotypic method that performs comparably with many of the most effective DNA-based methods. 21, 22 Characteristics of selected phenotypic methods are presented in Table 8.6. These methods have been characterized by: typeability, which is the ability of the technique to assign an unambiguous result (i.e., type) to each isolate; reproducibility, which is when a method yields the same results upon repeat testing of a bacterial strain; discriminatory power, which is the ability of the method to differentiate among epidemiologically unrelated isolates; ease of interpretation, which refers to the effort and experience required to obtain useful, reliable typing information using a particular method; and ease of performance, which reflects the cost of specialized reagents and equipment, technical complexity of the method, and the effort required to learn and implement the method.

TABLE 8.6. Characteristics of phenotypic typing methods

Typing System Proportion of Strains Typeable Reproducibility Discriminatory Power Ease of Interpretation Ease of Performance
Biotyping All Poor Poor Moderate Easy
Antimicrobial susceptibility patterns All Good Poor Easy Easy
Serotyping Most Good Fair Moderate Moderate
Bacteriophage or pyocin typing Some Good Fair Difficult Difficult
MLEE 1 All Excellent Excellent Moderate Moderate
1
MLEE, multilocus enzyme electrophoresis.

(modified from ref. 23)

Extremely sensitive and specific molecular techniques have recently been developed to facilitate epidemiologic studies. Our ability to use these molecular techniques (genotypic methods) to detect and characterize the genetic variability of infectious agents (bacteria, fungi, protozoa, viruses) is the foundation for the majority of molecular epidemiological studies. The application of appropriate molecular techniques has been an aid in the surveillance of infectious agents and in determining sources of infection. These molecular techniques can be used to study health and disease determinants in animal (including human) as well as plant populations. It requires choosing a molecular method(s) that is capable of discriminating genetic variants at different hierarchical levels, coupled with the selection of a region of nucleic acid, which is appropriate to the questions being asked (Table 8.7).

TABLE 8.7. Molecular characterization of genetic diversity at different hierarchical levels

Function Purpose Regions of DNA
Discrimination above level of species Taxonomy/evolution Highly conserved coding regions (e.g., rDNA)
Discrimination between species Taxonomy/diagnosis/epidemiology Moderately conserved regions
Discrimination between intraspecific variants/strains Population genetics Variable regions
Discrimination between individual isolates/clonal lineages "Fingerprinting"—tracking transmission of genotypes/identifying sources of infection and risk factors Highly variable genetic markers that are not under selection by the host
Genetic markers/linking phenotype and genotype Identifying phenotypic traits of clinical significance Genotype linked to phenotype

(modified from ref. 24)

Genotypic methods are those that are based on an analysis of the genetic structure of an organism. Over the past decade, a number of genotypic methods have been used to fingerprint pathogenic microorganisms (Table 8.8). The methods have been described in detail elsewhere. 23–27 Among these methods, RFLP-PFGE (restriction fragment length polymorphism/pulsed-field gel electrophoresis) and RFLP + probe, and ribotyping have been the most commonly used methods for fingerprinting bacteria. 25, 28 RAPD (random amplification of polymorphic DNA) and karyotyping have been used for fingerprinting fungi. 25, 29 MLEE (multilocus enzyme electrophoresis), RAPD, and PCR (polymerase chain reaction)-RFLP have been used for fingerprinting parasitic protozoa. 25 Select gene or complete genome characterization, as well as other molecular methods, have been used for viruses. 30

TABLE 8.8. Examples of genotypic methods used in epidemiologic investigations

Restriction endonuclease-based methods
A.

Restriction fragment length polymorphism (RFLP) without hybridization

—Frequent cutter (4–6bp recognition site) coupled with conventional electrophoresis to separate restriction fragments

—Infrequent cutter (generally 6–8 bp recognition site) coupled with pulsed-field gel electrophoresis (PFGE) to separate restriction fragments

B.

RFLP with hybridization

—Frequent cutter (4–6bp recognition site) coupled with conventional electrophoresis to separate restriction fragments followed by Southern transfer to nylon membrane. The power and efficacy of typing method depends on the probe.

—16S and 23S rRNA (ribotyping)

—Insertion sequences (e.g., IS6110 of Mycobacterium tuberculosis)

Amplification-based methods
A.

Random amplification of polymorphic DNA (RAPD) analysis; arbitrarily primed PCR (APPCR)

B.

Amplified fragment length polymorphism (AFLP) method

C.

Repetitive element PCR (REP-PCR) method; variable number tandem repeat (VNTR) fingerprinting
Sequence-based methods
A.

Multilocus sequence typing (MLST)

B.

Electrophoretic karyotyping

When should fingerprinting be used? Strain typing data are most effective when they are collected, analyzed, and integrated into the results of an epidemiological investigation. The epidemiologist should consult the laboratory when investigating a potential outbreak of an infectious disease. Microbial fingerprinting should supplement, and not replace, a carefully conducted epidemiological investigation. In some cases, typing data can effectively rule out an outbreak and thus avoid the need for an extensive epidemiological investigation. In other cases, these data may reveal the presence of outbreaks caused by more than one strain. Data interpretation is facilitated greatly by an appreciation of the molecular basis of genetic variability of the organism being typed and the technical factors that can affect results. With the exception of whole-genome sequencing, the molecular methods analyze only a small portion of the organisms' genetic complement. Thus, isolates that give identical results are classified as "indistinguishable," not "identical." Theoretically, a more detailed analysis should uncover differences in the isolates that appeared to give identical patterns, but that were epidemiologically unrelated. This is unlikely to occur when a set of epidemiologically linked isolates are analyzed. 23 For this reason, only whole-genome sequencing would provide the unequivocal data required for attribution.

The power of molecular techniques in epidemiological investigations is well exemplified by a few examples. PulseNet, the national molecular subtyping network for foodborne disease surveillance, was established by the CDC and several state health departments in 1996 to facilitate subtyping bacterial foodborne pathogens for epidemiologic purposes. Twenty years ago, most foodborne outbreaks were local problems that typically resulted from improper food-handling practices. Outbreaks were often associated with individual restaurants or social events, and often came to the attention of local public health officials through calls from affected persons. Today, foodborne disease outbreaks commonly involve widely distributed food products that are contaminated before distribution, resulting in cases that are spread over several states or countries. The PulseNet network, which began with 10 laboratories typing a single pathogen (Escherichia coli O157:H7), has grown and now includes 46 state and two local public health laboratories and the food safety laboratories of the U.S. Food and Drug Administration (FDA) and the U.S. Department of Agriculture (USDA). 28 Currently, four foodborne pathogens (E. coli O157:H7, nontyphoidal Salmonella serotypes, Listeria monocytogenes, and Shigella) are being subtyped by PFGE as part of routine surveillance for foodborne disease. The laboratories follow a standardized protocol using similar equipment so that results are highly reproducible and DNA patterns generated at different laboratories can be compared. Isolates are subtyped on a routine basis, and the data analyzed promptly at the local level. Clusters can often be detected locally that could not have been identified by traditional epidemiologic methods alone. PFGE patterns are shared between participating laboratories electronically, which serves to link apparently unrelated outbreaks and facilitates the identification of a common vehicle. 31 For example, in May 1998, PulseNet facilitated the investigation of two clusters of E. coli O157:H7 in the northeastern U.S. PFGE fingerprinting of the E. coli O157:H7 isolates by the PulseNet laboratories in that region revealed two simultaneous clusters of E. coli O157:H7 infections (32 isolates in four of five states with one PFGE pattern, and 25 isolates in all five states with a second pattern), one of which could be traced to two supermarkets that received ground beef from the same distributor. Without molecular typing, epidemiologists would have found it difficult to identify cases associated with each cluster. On the other hand, the use of PFGE subtyping as part of routine surveillance has benefits beyond outbreak detection. For example, the temporal clustering of unrelated cases is not uncommon, and without molecular typing, valuable public health resources would be wasted investigating pseudo-outbreaks.

Another example of the power of molecular techniques in solving an epidemiologic investigation involves a case of HIV transmission by a healthcare worker. The investigation involved a young woman who had contracted AIDS even though she had no identifiable risk factors. During the investigation, it was revealed that 2 years previously she had several teeth extracted by a dentist who was subsequently confirmed as having AIDS. A retrospective case-control study was conducted of the dentist and his former patients to evaluate the possibility of dentist-to-patient transmission. Patients were questioned to ascertain known risk factors for HIV transmission. Infection control practices in the dental office were also evaluated. 32 Eight HIV-positive persons were identified from among a group of more than 1,000 former patients of the dentist. Five of the eight patients had no risk factors or other documented exposures to HIV. Although all five had undergone invasive procedures, and four of the five shared visit days, no identifiable mechanism of transmission could be established by traditional case-control methodology. 32 However, a comparison of the nucleotide sequences of several regions of the gp120 gene of the HIV strains of the dentist, HIV-positive patients (with and without known risk factors), and 35 HIV-infected community controls established the likelihood of a common source of infection. 33 The genetic distance of viruses from the five patients without known risk factors and the virus from the dentist was 3.4%–4.9%, which is similar to that found previously with HIV viruses from persons with epidemiologically linked infections. In contrast, isolates from patients with known risk factors were more distantly related (>10%) to the HIV virus obtained from the dentist. 33 The average genetic distance of viruses from the five patients and the community controls was approximately 11%, which was virtually identical to the average distance among the 35 HIV viruses from controls. Phylogenetic tree analysis confirmed that the HIV viruses from the dentist and the five patients formed a tightly related cluster. 33

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780120884834500111

Epidemiology of infectious diseases in immigrants

Fernando Cobo , in Imported Infectious Diseases, 2014

2.4.2 Prevention and control programs

It is well know that patterns of immigration and the epidemiology of infectious diseases can change quickly. Thus, there is a need to assess prevention and control policies and programmes with respect to these challenges. However, traditional prevention and control programmes have not been sufficient for immigrant populations living in developed countries. Vaccination programmes in these populations should be improved in order to decrease preventable diseases.

There is little evidence that screening programmes are useful; where immigrant screening programmes are available, there is some evidence that the tests are not always sensitive. The efficacy of any prevention and control programme depends on disease prevalence, the available test methods, their sensitivity and predictive value, and the existence of relevant healthcare measures.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9781907568572500027

Models for the Study of Infection in Populations

John R. Williams , in Handbook of Models for Human Aging, 2006

MODEL VALIDATION AND SENSITIVITY AND UNCERTAINTY ANALYSIS

An essential aspect of the use of mathematical models in infectious disease epidemiology is validation of model results against real data, validation here referring simply to the ability to satisfy oneself that the model results are consistent with the available data relating to the population which is being modeled (or in its absence, data from a population sharing similar characteristics). It is not necessary for this process to be one of fitting, in the statistical sense, as we are dealing with model results which are qualitative rather than quantitative, and it is usually therefore more important that the model can reproduce the change in shape over time of the data being used for validation rather than absolute values. Thus the process is more one of inspection and fitting by eye rather than statistical fitting. Given that we can establish the validity of the model results in this way, it is important to know how sensitive these results are to the values of the model parameters. Depending upon the structure of the model and the role played by each parameter in the model, very small changes in the values of some parameters can lead to large variations in model results; for other parameters the situation may be reversed and relatively large changes in values may have little impact on the results. It is the role of sensitivity analysis to establish how variation in parameter values might impact on model outputs; if the model should prove to be too sensitive, the results will be useless. This can either be done simply by varying the values of each parameter in turn and observing what effect each has, or more systematically by the use of "Latin hypercube" sampling (cf. Latin square), which involves specifying a distribution for the values of each parameter (which could simply be a uniform distribution if the true distribution is not known) and sampling at random and without replacement parameter values from the combined set of distributions (Seaholm et al., 1988). Once the sensitivity of the model to its parameters has been established, it is necessary to consider how much uncertainty there is in our knowledge about the true values of each parameter in relation to the population being considered. In uncertainty analysis we are considering what the plausible range of values for each parameter might be, and exploring the variation in model results which arises when parameter values are varied within this set of ranges; the variation in model outputs then provides an indication of uncertainty about the likely way in which, for example, prevalence or incidence of an infection may change over time in the population being studied.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780123693914500163

Computational Modeling in Global Infectious Disease Epidemiology

Ali Alawieh , ... Fadi A. Zaraket , in Leveraging Biomedical and Healthcare Data, 2019

4 Conclusions

In this chapter, we have described two novel applications of computational modeling in the field of infectious disease epidemiology that study disease-specific epidemiology at international or national levels. In addition to the approaches described here, there has been significant advances in the application of computational modeling in infectious disease epidemiology over the past decade that included, among others, the study of spatial transmission of disease and population migration (e.g., GLEaM model ( Balcan et al., 2010), and influenza propagation in gatherings (Shi et al., 2010)), the study of disease transmission in hospital settings (Cooper and Lipsitch, 2004; Donker et al., 2010), the study of the impact of vaccination (Burke et al., 2006; Duintjer Tebbens et al., 2008b), and the study of pathophysiological features impacting infectious disease progression (Castiglione et al., 2004).

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780128095560000095