A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
(The study of) the causes of disease.
The inherited potential to resist a disease or infection. See also immunity.
The reduction in disease when a risk factor is removed. If I* is the number of people that a risk factor is responsible for and I is the total number of cases over the same period, then the aetiological fraction is I*/I. Equivalently, the aetiological fraction is (I-Io)/I. where Io is the number of cases in the absence of the risk factor. Also known as the attributable fraction.
A mathematical model which take into consideration the division of the host population into different age classes. Such models can used to consider the consequences of such factors as age-dependent infection, morbidity or mortality rates or of age-specific vaccination schedules.
A distribution of parasites amongst hosts is said to be aggregated, or over dispersed, if parasites are found to co-occur in particular hosts more often than if the parasites were distributed at random amongst all hosts. A more precise test would be if the distribution differed significantly from a Poisson distribution; a good heuristic is whether the variance to mean ratio is bigger than unity (in appropriate units).
Macroparasites are typically aggregated in their host populations, so that the majority of hosts harbor few or no parasites while a few hosts harbor large parasite burdens. Aggregated distributions are often well described empirically by the negative binomial distribution.
A focused study of the determinants of disease or reasons for high or low frequency of disease in specific groups.
The period between conception and birth. Same as prenatal.
A protein produced in the blood of vertebrates following exposure to an antigen. The antibody binds specifically to the antigen and thus stimulates its inactivation by other parts of the immune system.
The major classes of antibody are immunoglobulin A, or IgA, found predominantly in bodily secretions like saliva; IgM and IgG which are typically produced sequentially in response to microparasitic infections; and IgE which is often elevated in the response to helminth infections. Only IgG is able to cross the placenta to provide maternal immunity.
A protein, typically foreign, that elicits a specific immune response.
A drug, for example praziquantal, used specifically against helminth or worm infections.
A virus which uses Arthropods as vectors and is transmitted in their saliva to the definitive host. For example, yellow fever.
(From Arthropod borne virus.)
Artifactual (spurious) Association
A false association that occurs by chance or through bias.
The same as aetiological fraction.
A white blood cell which produces antibody.
The presence of bacteria in the blood.
See Reproductive Ratio.
Birth, Immigration, Death, Emigration: the four demographic processes which might act on a population compartment in a typical compartmental model.
See transmission breakpoint.
An individual who is infected but has no symptoms of disease.
There are two types of carrier state: silent carriers retain their infectiousness, while latent carriers are not infectious. For example, many of those infected with tuberculosis are silent carriers, while infection with herpes virus may create latent carriers.
The maximum number of individuals a habitat can sustainably support. Typically defined so as not to include losses from predation or disease.
A (rather misleading name for a) type of compartmental model in which the force of infection is treated as a parameter to be estimated.
A study in which the risk factors of people with a disease are compared with those without a disease.
Drug treatment of a diseased individual.
Drug treatment designed to prevent future occurrences of disease. Treatment may be chemotherapy as far as an individual is concerned but chemopropylactic for the population as a whole.
Our term for the varieties of epidemiology primarily concerned with the statistical relationships between disease agents, both infectious and non-infectious; for example a study to establish the relative risk of lung cancer associated with smoking.
We contrast this with ecological epidemiology.
An infection which requires close contact, other than sexual contact, between susceptible and infectious individuals, for transmission.
Cytomegalovirus: a herpes virus which causes glandular fever (as does EBV).
Changes in the genotypes of two or more species that are a direct consequence of the species' interaction with one another.
A subsection of a population with a common feature, usually age. For example, all those individuals in the UK born in 1964 form a birth cohort.
A mathematical model which divides hosts into different compartments according to their infectious state. A typical model for microparasites might be an SEIR model. Sometimes referred to as a prevalence model.
The rate at which susceptibles meet infecteds. Usually measured as individuals per unit time.
Same as an aggregated distribution.
A model in which the system changes continuously over time. Derivatives (e.g. dY/dt) are the mathematical formalism for describing such continuous change. The differential equation which embodies a model provides the values of these derivatives at any particular time point; calculus or a computer can then be used to move the state of the model forwards in time.
Continuous models have the advantage over discrete time models in that they are more amenable to algebraic manipulation, although they are slightly harder to implement on a computer.
The same as a differential equation model.
Congenital rubella syndrome.
Same as a horizontal study.
The number of live births in a year divided by the population size.
The number of deaths in a year divided by the population size.
The host in which a parasite reproduces sexually.
1) Effects whose intensity changes with increasing population density.
2) Effects whose intensity increases with increasing population density, for example a mortality which becomes very high near a carrying capacity. Effects which become less pronounced at higher population sizes (perhaps the reduction in fecundity caused by the difficulty of mating) are sometimes called inverse density dependent. Density dependent effects are often a cause of nonlinearity in a model.
A mathematical model in which the parameters and variables are not subject to random fluctuations, so that the system is at any time entirely defined by the initial conditions chosen. Contrast with a stochastic model.
A device for starting futile arguments over definitions.
The mathematical formulation corresponding to a continuous model; an equation involving derivatives.
The mathematical formulation corresponding to a discrete time model.
Having separate sexes, opposite of monoecious.
A life cycle in which a parasite is transmitted directly from one host to the next without an intermediate host or vector of another species.
A model in which the system jumps from one state to the next at fixed intervals or timesteps. These difference models are simple to understand but often difficult to analyze; Contrast continuous time models.
The parameters in such a model refer to the amount of change over the finite timestep; they are sometimes referred to as finite rates.
In a (rather precise) sense, a differential equation is what you eventually get from a difference equation when you let the timestep get smaller and smaller and smaller.
The debilitating effects on a host of infection by a parasite.
The pattern by which parasite numbers are partitioned amongst available hosts. If the fraction of the population who have i parasites is p(i) then the distribution is the set of numbers p(0),p(1), p(2), ... . Summary statistics of this distribution include prevalence (1-p(0)) and mean intensity ( p(1)+ 2p(2)+ 3p(3)+...). Useful distributions include the Poisson and negative binomial.
Epstein-Barr virus: a herpes virus which causes glandular fever (as does CMV) and some cancers.
A branch of epidemiology which views disease as a result of the ecological interactions between populations of hosts and parasites. We contrast this with classical epidemiology.
An index of the potency of a drug or disease treatment.
For an Anthelmintic, usually estimated as the proportion of the worm burden in a host killed by a single dose or short-term course of the treatment.
Efficacy is the measure of the impact of a treatment e.g. vaccine, under trial conditions (as opposed to effectiveness which is its impact within the population). Thus efficacy is the percentage reduction in infection or disease caused by a vaccine in a trial group compared to a control group. Specifically, if Iv is the incidence in vaccinated individuals and Iu is the incidence that would have occurred in that group had they not been vaccinated, then vaccine efficacy for that group is v=1-Iv/Iu.
Note that this may include the effects of protection due to
herd immunity, and so is not the same as same as immunogenicity,
even if the presence of an immune reaction always indicated individual
A technique using the antigen binding properties of antibodies to detect specific antigens or antibodies. Visualization is typically made possible by enzyme induced color formation.
From Enzyme Linked ImmunoSorbent Assay.
A term to describe levels of infection which do not exhibit wide fluctuations through time in a defined place.
For microparasites like measles, the term is used slightly differently to indicate an infection which can persist in a population in the long term without needing to be reintroduced from outside.
Stable endemicity is where the incidence of infection or disease shows no secular trend for increase or decrease.
Parasite extinction occurring because endemic levels are so low that it is possible for small stochastic fluctuations to remove all parasites. Contrast epidemic fadeout.
A rapid increase in the levels of an infection. Typical of the microparasitic infections (with long lasting immunity and short generation times) an epidemic is usually heralded by an exponential rise in the number of cases in time and a subsequent decline as susceptible numbers are exhausted. Epidemics may arise from the introduction of a novel pathogen (or strain) to a previously unexposed (naive) population or as a result of the regrowth of susceptible numbers some time after a previous epidemic due to the same infectious agent. Contrast endemic, pandemic.
Parasite extinction occurring because numbers are so low immediately following an epidemic that it is possible for small stochastic fluctuations to remove all parasites. Contrast endemic fadeout.
The study of the distribution and determinants of diseases and injuries in human (or animal) host populations.
An epidemic in an animal host population. (But while there may be scientific reasons for distinguishing a human epidemic from an epizootic, the use of epizootic is not required on linguistic grounds any more than a demographer is required to count only people: see Nature, 368, 284, (1994))
An epidemic in a plant host population.
A state in which a system is not changing. A population size might be at a static equilibrium at which nothing is happening (there are no births or deaths) or a dynamic equilibrium at which different processes are balanced (there are the same numbers of births and deaths). More generally, the state to which a system eventually evolves, for example sustained periodic oscillations, might be called an equilibrium.
A decline in which the rate of decay is always proportional to the amount of material remaining; the constant of proportionality is the rate constant.
An increase in which the rate of growth is always proportional to the amount of material remaining; the constant of proportionality is the rate constant.
See Odds ratio.
The capacity to produce offspring; the rate of offspring production.
The per capita rate at which susceptibles are infected.
Hepatitis A, B and C virus. These very different viruses all cause the liver disease hepatitis. Hepatitis B and C are blood borne, while Hepatitis A is an enterovirus which is fecal-orally transmitted. Other known hepatitis viruses include Hepatitis D and Hepatitis E.
Worms, in particular the five groups of parasitic worms: Monogeneans (flukes), Digeneans (flukes, schistosomes), Cestodes (tapeworms), Nematodes (roundworms) and Acanthocephalans (spiny-headed worms).
Specifically, the mechanism by which an infection may be eradicated from a population although some susceptibles still remain, because the remainder of the population is immune and thus transmission is reduced.
More generally, the immunological status of a population of hosts and its effect on transmission rates.
Hemagglutination inhibition test, a serological test used to detect antibodies specific to a particular family of viruses which possess the ability to agglutinate red blood cells e.g. measles, rubella and influenza.
Haemophilus influenzae B.
Human immunodeficiency virus: the cause of Acquired Immune Deficiency Syndrome (AIDS).
Human papillomavirus, causing genital warts and genital cancers.
Herpes simplex virus. Types one and two, HSV-I and HSV-II, are among the causes of cold sores and genital ulcers.
An infection whose prevalence is fairly uniform throughout a region, country or continent.
Mainly used in the malaria literature.
A study of a community, perhaps stratified by age, sex, ethnicity etc., but at one point in time or over a short time interval.
Although a snapshot, horizontal surveys of prevalence and intensity within different age classes of a community can nevertheless provide valuable information on the rate at which hosts acquire infection through time, provided that the host and parasite populations have remained approximately stable for a period of time (i.e. stable endemicity). Same as cross-sectional study, converse of longitudinal study.
Transmission occurring generally within a population, but not including vertical transmission.
A term from the malaria literature used to mean (roughly) an area with little transmission.
1) A state in which a host is not susceptible to infection or disease, or
2) the mechanisms by which this is achieved. Immunity is achieved by an individual through one of three routes: natural or innate immunity genetically inherited or acquired through maternal antibody, acquired immunity conferred after contact with a disease, and artificial immunity after a successful vaccination. Also termed specific immunity, resistance or specific resistance.
Specific immunity is divided into cellular immunity acting via the direct involvement of T cells and humoral immunity involving antibodies and B cells.
The ability of a vaccine to stimulate the immune system, as measured by the proportion of individuals who produce specific antibody or T cells, or the amount of antibody produced, say. Not the same as efficacy.
Damage to the host caused by its own immune response against a pathogen.
A reduction in the capacity of the immune system. Caused by infection (e.g. HIV), drug treatment, pregnancy and malnutrition among others. Immunosuppressed individuals are commonly referred to as immunocompromised.
The rate at which new cases of infection arise in a population. Classically measured as an attack rate.
The time that elapses between infection and the appearance of symptoms of a disease. Not the same as the latent period.
A life cycle which requires one or more intermediate hosts before the definitive host species is reinfected. Compare direct, nondirect.
Transmission of a parasite through an indirect life cycle.
The presence of a parasite within a host where it may or may not cause disease.
n., A host who has an infection.
The time period during which infecteds are able to transmit an infection to any susceptible host or vector they contact. Note that the infectious period may not necessarily be associated with symptoms of the disease.
The amount of parasite to which an individual host is exposed at transmission.
In a short time interval (e.g. a week), the number of events (e.g. births in the UK) taking place during the interval is approximately proportional to the length of the interval. The constant of proportionality is the rate at which these events occur. This argument becomes more and more exact as the length of the time interval becomes shorter and shorter. When the time interval has shrunk to an instant, the rate has become an `instantaneous rate'. A bad term; just call it a rate.
1) (Traditional) The mean parasite burden within all the infected members of the host population. Also called mean abundance.
2) (Newer) The mean parasite burden within both infected and uninfected hosts.
It is important to indicate which usage is adopted, since they give different statistics, unless the prevalence is 100%. Macroparasites, and infections like malaria, are usually measured in terms of intensity.
See density dependence.
The time from infection to when the individual is infectious to others. In helminths it is termed the pre-patent period. Not the same as the incubation period.
Longevity, the average length of life of individuals in a population.
Given some data (for example maternal antibody levels), and a model for how a process supposedly generating it occurs, (for example, exponential decay with rate m ), and for how variance is likely to occur in observations of the process (for example sampling error), the likelihood that a particular model (i.e. a particular m plus the error in observation can generate the observed data. Picking the parameter value which is most likely to have generated the observed data is the method of maximum likelihood.
A process is linear if doubling the starting conditions doubles the outcome. For example, the number of new schistosome infections establishing in mice is directly proportional to the number of cercaria the mice are exposed to, over a certain range. Outside this range, the response is nonlinear: there is no longer a direct proportionality.
Linear differential equations are ones whose solutions are linear. This property allows us to solve linear differential equations completely: most nonlinear differential equations can't be solved analytically.
A study taking place over time. If individuals are followed, this is a longitudinal cohort study. If individuals are not followed, but classes (usually age classes) are restudied, this is a longitudinal cross-sectional study.
The converse of a horizontal study.
Transmission of infection which occurs at a rate directly proportional to the number or density of both susceptibles and infecteds present.
Some authors reserve the name mass action for transmission processes of the form b X Y/N, which we associate with STD-type transmission, and describe transmission rates of the form b X Y, as pseudo-mass action; the two are equivalent if the population size is unchanging.
Typically, the parasitic helminths and arthropods. In general, parasites which do not multiply within their definitive hosts but instead produce transmission stages (eggs and larvae) which pass into the external environment. Immune responses elicited against macroparasites generally depend on the number of parasites present in a given host and tends to be of a relatively transient nature. The key epidemiological measurement is generally the number of parasites per host.
Such parasites are often found in a highly aggregated distribution.
Immunity for a neonate provided by IgG antibody generated by a mother and passed across the placenta to the unborn offspring. This provides short lived protection (with a typical half life of 3-6 months) to the neonate. Also known as immunity.
A formal framework to convey ideas about the components of a host-parasite interaction. Construction requires three major types of information: (a) a clear understanding of the interaction within the individual host between the infectious agent and the host, (b) the mode and rate of transmission between individuals, and (c) host population characteristics such as demography and behavior.
Mathematical models can aid exploration of the behavior of the system under various conditions from which to determine the dominant factors generating observed patterns and phenomena. They also aid data collection and interpretation and parameter estimation, and provide tools for identifying possible approaches to control and for assessing the potential impact of different intervention measures.
A term from the malaria literature used to mean (roughly) an area with some transmission.
Typically, viruses, bacteria, fungi and protozoa. More generally, parasites which multiply within their definitive hosts. Microparasites are characterized by small size, short generation times, and a tendency to induce immunity to reinfection in those hosts that survive. The duration of infection is usually short in relation to the lifespan of the host, but there are important exceptions, such as the slow viruses. The key epidemiological variable, by contrast with macroparasites, is whether or not the individual host is infected.
Chemical substances which kill snails or other mollusks.
Male and female sexual apparatus present in the same individual. Same as hermaphrodite; opposite of dioecious.
State of ill-health produced by a disease.
The per capita death rate in a population. The mortality rate is the reciprocal of the population life expectancy.
An infection in which an individual is infected by parasites of more than one species.
A distribution which is parameterized by a mean m and an aggregation parameter k which is large when aggregation is small; in fact as k becomes large, the negative binomial distribution approximates the Poisson distribution.
Diseases, usually of an infectious nature, whose occurrence is required by law to be made known to a health officer or local government authority.
A mode of transmission that differs in some mysterious way from indirect transmission; we don't make a distinction in this course.
A comparison of the presence of a risk factor for disease in a sample of diseased subjects and non diseased controls. The number of people with disease who were exposed to a risk factor (I.e.) over those with disease who were not exposed (Io) divided by those without disease who were exposed (Ne) over those without who were not exposed (No). Thus OR=(I.e./Io)/(Ne/No)= I.e. No / Io Ne. This measure should be used for case control studies where we retrospectively look at risks in those with and without disease. Also known as exposure odds ratio.
An epidemic widely distributed in space.
A widely distributed epizootic, often one affecting more than one host species.
Immunity which has been acquired through the transfer of maternal or other antibodies. Passive immunization does not induce immunological memory.
1) Any disease causing organism. 2) An organism exhibiting an obligatory dependence on another organism, its host, which is detrimental to the host.
Same as a parasite.
The degree to which a pathogen debilitates its host.
Between the 28th week of pregnancy and the end of the first week of life.
A rate which is proportional to the number of individuals in a population.
The distribution which arises when parasites are distributed at random amongst hosts.
Subsequent to (and within 1 year of) childbirth.
The time from infection until when a female starts to produce eggs in helminth infections, equivalent to latent period in micro-parasitic infections.
An animal that kills its prey and then feeds on it to subsist until the next kill.
The proportion of the host population infected (or with some marker of past or present infection) at a particular time.
Prevalence models are compartmental models dividing the host population into, for example, susceptible, latent, infectious and immune individuals.
A study in which people are initially enrolled and then followed up at subsequent times.
A unicellular eukaryote.
A period during which an infection is present but not active within a host: for example the period between an acute attack of chickenpox (varicella) and a subsequent recrudescence of shingles (zoster); not the same as latency.
The number of events happening divided by the length of time over which they happen. A rate of change is the amount of change happening in a interval divided by the length of the interval; for small intervals the rate of change might be given by a simple rule (a differential equation).
Reappearance of disease in a host whose infection has been quiescent.
(1) Basic reproductive ratio, Ro, a dimensionless parameter which encapsulates the biological details of different transmission mechanisms. For microparasites, Ro, is defined as the average number of secondary cases of infection to which one primary case gives rise throughout its infectious period if introduced into a defined population consisting solely of susceptible individuals. For macroparasites, Ro, is the average number of female offspring (or just offspring in the case of hermaphroditic species) produced throughout the lifetime of a mature female parasite, which themselves achieve reproductive maturity in the absence of density-dependent constraints on the parasite establishment, survival or reproduction.
Also known as the basic reproduction rate, number.
(2) Effective reproductive ratio, R, The number of secondary cases (microparasites) or female offspring (macroparasites) produced in a host population not consisting entirely of susceptible individuals (microparasites) or within which density dependent constraints limit parasite population growth (macroparasites). Under conditions of stable endemic infection, R=1.
The proportion of diseased people amongst those exposed to the relevant risk factor divided by the proportion of diseased people amongst those not exposed to the risk factor. This should be used in those cohort studies where those with and without disease are followed to observe which individuals become diseased.
1) The reduction, due to genetic selection, of susceptibility of a parasite or its vector to chemotherapy.
2) The ability of a host to resist a pathogen. Compare immunity.
A study in which people are enrolled and then have their history of risks, infections or disease measured.
A class of compartmental prevalence models, with compartments Susceptible, Latent (Exposed), Infectious and Recovered. Takes its name from a common notation. In the notation used in the course, this would be an XHYZ model. Many permutations possible.
The ability of a test to work on people you know have the infection. More precisely TP/(TP+FN), where TP is the number of true positives and FN is the number of false negatives
The study of antigen-antibody reactions. More generally, the use of serotype data to infer an individual's history of infection.
n., an individual whose serotype suggests that they have experienced infection in the past.
The proportion of a population who are seropositive.
The range of antibodies which an individual possesses, usually based on sampling from blood serum or saliva.
Different strains of a pathogen can sometimes be distinguished by the different antibodies they induce in a host, or with which they can be made to react in vitro; thus the word serotype has also come to be applied to a particular strain (`the virulent serotype'). This is the more common clinical usage.
The range of antibodies used to define a serotype obviously depends on those available to the researcher. Sometimes, as for measles, the presence of a known antibody within the serum of an individual correlates extremely well with the clinical observation that that individual is protected against any further infections. But sometimes, as for malaria, there is as yet no definite relationship between a given serotype and the presence of a functional immunity, which may make the use of the word serotype unhelpful when trying to distinguish between different parasites for the purposes of understanding transmission.
The ability of a test to fail on people you know don't have the infection. More precisely TN/(TN+FP), where TN is the number of true negatives and FP is the number of false positives
A mathematical model which takes into consideration the presence of some randomness in one or more of its parameters or variables. The predictions of the model therefore do not give a single point estimate but a probability distribution of possible estimates. Contrast with deterministic.
We might distinguish demographic stochasticity which arises from the discreteness of individuals and individual events such as birth, and environmental stochasticity arising from more-or-less unpredictable interactions with the outside world.
An infection in which symptoms are sufficiently mild or unapparent to escape diagnosis other than by positive confirmation of the ability to transmit the infection or serologically.
Sexually transmitted disease.
An individual accessible to or liable to infection by a pathogen.
A condition of the body reported by an individual when suffering from a disease; here used more loosely to include signs: any evidence used in diagnosis or identification of infected individuals.
A white blood cell responsible for cell-mediated immunity in particular and immunoregulation in general.
The number of children an average women would have assuming that she lives her full reproductive lifetime.
The process by which a pathogen passes from a source of infection to a new host. There are two major types: horizontal and vertical transmission. The majority of transmission processes operate horizontally.
A critical average worm burden below which the mating frequency is too low to maintain a dioecious parasite species.
Occurs for a parasite when the basic reproductive rate Ro is equal to 1. Below this threshold level the disease is unable to maintain itself within the host population.
A mortality schedule in which all hosts are assumed to live for a fixed number of years equal to the life expectancy.
A mortality schedule in which all hosts are assumed to die at a constant rate. This constant rate is equal to the inverse of the life expectancy.
A drug intended to induce active artificial immunity against a pathogen. Vaccines may be live or dead. Live vaccines are usually attenuated versions of the wildtype pathogen, such as the MMR vaccines, which are strains of measles mumps and rubella viruses repeatedly passed through cell lines until non-pathogenic. Typically, live vaccines need only be given as a single dose to induce a full immunological response, inducing specific memory. Dead vaccines are either killed whole parasite, as with the Salk polio strain and pertussis vaccine, or some highly immunogenic fraction of the parasite, as in toxoid vaccines. Killed vaccines and toxoids which do not multiply in the host must usually be administered in multiple doses to induce a full immunological response.
Vaccination should be distinguished from passive immunization in which concentrated specific antibodies which can be used therapeutically to abrogate an ongoing infection or to provide short term protection (of the order of months), for example against hepatitis A. Passive immunization does not induce immunological memory.
1) The intermediate hosts of parasites with indirect life cycles.
2) Anything which transmits parasites.
In vector-borne infections such as malaria, the vectorial capacity is a concept analogous to the contact rate in directly-transmitted diseases. It is, thus, a function of (a) the vector's density in relation to its vertebrate host, (b) the frequency with which it takes blood meals on the host species, (c) the duration of the latent period in the vector, and (d) the vector's life expectancy.
Vertical transmission occurs when a parent conveys an infection to its unborn offspring, as occurs in syphilis in man or in many arboviruses of arthropods. Perinatal infection is a special form of vertical transmission.
The presence of virus in the blood.
A mature and infectious virus particle.
(1) The case mortality rate of an infection. (2) The extent to which a pathogen harms its host. These are different usages: what they have in common is that they refer to the effect on an already infected host, not to the degree of transmissibility to a subsequent susceptible.
Varicella-zoster virus: a herpes virus which causes chickenpox (varicella) and shingles (herpes-zoster).
The number of worms an individual host carries. Sometimes directly measurable (Ascaris), sometimes only indirectly (schistosomiasis).
A parasite naturally transmitted between man and other vertebrate species
Source: Jonathan Swinton
This dictionary was edited by Jonathan Swinton, but bears traces of descent from one written by Andy Dobson, with many suggestions, additions, improvements and removals by Graham Medley, James Nokes, Geoff Garnett, Bryan Grenfell, Catherine Watt, Andrew Hall and Michael Clarkson. A more pure-bred descendant is the glossary by Watt, Dobson and Grenfell in Grenfell and Dobson, Ecology of infectious diseases in natural populations, Cambridge University Press, 1995. Last, A Dictionary of Epidemiology, Oxford University Press, 1995 provides an authoritative printed dictionary of classical epidemiology, with some cross coverage of these terms.