Topic 1 Quantitative Indices in Health and Disease (The Use of Rates and Ratios and Basic Statistics)
Topic 2 Temporal Distributions ( Clustering patterns by Time)
Topic 3 Spatial Distributions ( Clustering patterns by Space or Place) and Host Related Distributions ( Clustering patterns by Host)
Chapters 2 , 6, 7 - Textbook
Objective: To understand the effect of time as a factor in determining the epidemiologic study of infections.
Problem Definition: How does one decipher patterns of occurrence of diseases or health conditions in populations?
Introduction: Describing diseases in populations
This phase of epidemiology deals with observing, describing and recording the occurrence and evolution of disordered states of health in a population and seeks to discover their causes and prevent them. For these purposes, enumeration is an essential first step. It is carried out by answering a number of questions about the disease, the first of which is:
How much disease (cases) is (are) there in a population?
It is answered by standard methods of disease accounting. The next question is:
when do these cases occur, i.e. what is their temporal distribution?
Various methods are used to describe disease in populations as it occurs over time; a particularly useful one is the epidemic curve.
The next question is:
where do they occur?
This question is approached by identifying the spatial distribution of cases. The approach relies on available information about the physical (geographic) location of cases and uses a number of mapping techniques. A fourth question one needs to answer is:
who are the cases and the non-cases?
"Who" refers to the specific characteristics such as breed, age, sex, etc. of the animals that are (or are not) cases. These characteristics are also termed host factors, the animal being the host of the disease agent whether the latter is a microorganism, an environmental toxicant, or a trauma victim resulting from inappropriate management practices. Additional factors (descriptors) that need to be considered concern the environment, such as weather, husbandry practices, etc.
Important among these are specific agents of disease. Physical and biological changes in such agents may influence the occurrence and distribution of disease.
The rationale for approaching the study of diseases in populations in the manner described in the preceding paragraphs may be found in two premises of modern epidemiology. These are:
1. that diseases in populations do not occur in random fashion, and
2. that diseases in populations do have multiple determinants.
Topic 1: Quantitative Indices of Health and Disease in Populations
Accounting of disease events--The use of rates and ratios:
An integral part of descriptive epidemiology is disease accounting. For example, disease accounting is one of the first activities in an outbreak investigation. It is important that both cases as well as non-cases are accounted for. A "case" will be defined in each situation, e.g. it may be an affected animal or one that merely has an antibody titer to a given disease agent. The "non-case" may be an animal that could become a case (an animal at risk) or an animal that was a case and has recovered. Both of these categories of animals will be counted, the cases usually receiving more attention than the non-cases. However, it is important not to neglect the latter ones since they contribute to the denominator (population at risk) of some very important rates.
In epidemiology, events such as disease or deaths are expressed in terms of rates or proportions. The reason for this is that numeration of cases makes sense only when the resulting numbers or numerator is related to a population at risk, i.e. the denominator. Furthermore, a rate implies a time element, i.e. the frequency of events occurring in a population is measured over an interval of time. If this interval is "long" the size of the population at risk may change, hence one uses an average population for that interval, and while it may be relatively easy to count the number of cases occurring over time, establishing the denominator requires care.
A. Rates, ratios, proportions.
A rate is used in epidemiology to measure the frequency of events (cases) as they occur over time. Rates express the probability of occurrence of some particular event.
A rate is of the form:
r is the number of cases that occur during a specified interval of time;
n is the total number of animals in the population at risk of becoming cases during the same interval;
c is a constant such as 100, 1000, etc. Note that r (the numerator) is a portion of n (the denominator).
A proportion is of the same form as a rate; (c being equal to 100 or % since proportions are usually expressed as percent - %).
An index is an estimate of a rate and is used when it is impossible to count directly the population at risk (n). The denominator is then obtained by counting some parameter which approximates the population at risk.
A ratio (k) expresses the relationship between a numerator and a denominator that does not include the numerator. For example:
* k is a ratio
* a is the numerator
* b is the demominator (does not necessarily include the numerator)
The two categories of rates most commonly used are:
1. those measuring disease events (morbidity rates), and
2. those measuring deaths (mortality rates).
Other rates are used to measure recovery, reproduction (e.g. birth rate), and contact between animals. Some of these important rates are summarized in the Table.