Healthcare and Medicine Reference
being treated by drugs, having surgery, having complications, suffering from
adverse effects of treatment and care, etc.) in various groups of individuals.
Events are presented and studied in two forms: frequencies or fractions.
Frequencies are simple numbers of events happening. Example: 250 cases
of chronic obstructive pulmonary disease (COPD) exist in our community. It is a
total of events as they were observed. Frequencies such as 1000 cases of cancer
or their rates (1,000 cases per 100,000 subjects in the community; i.e., a rate of
1/100 or 10/1,000, etc.) represent absolute values of what we observed, usually a
total of events as they were observed in a given category, set, or group.
Fractions are two related frequencies as expressions of relationship
between them. Frequencies of events are numerators presented as parts of
some whole. The whole is presented as the denominator. Example: 25 cases
of COPD occur in a community of 35,000 inhabitants (25 diseased + 34,775
nondiseased). In epidemiology, they are usually presented as rates or ratios .
Fractions are expressions of the relationship between two entities. In our
case, these expressions are most often rates (proportions) or ratios.
Rates are couplings of a set of observations in the numerator like disease
cases (a) with a set in the denominator including those same observations
(a) and some additional ones such as noncases (b) such as two sets creat-
ing a community: a / a + b . A rate in our example is then 25/(25 + 34,775). It
can be presented as such or multiplied by some coefficient like 100,000 for
comparability: 25/35,000 × 100,000, that is, 7142.85/100,000. For less frequent
phenomena like cancer, rates per 100,000 may be presented. For frequent
phenomena like diarrhea cases during a food born disease outbreak per
100, adverse drug effects may be reported per 1,000,000, and so on. As in
our COPD case, rates may also be presented as a percentage (07.14%) or as
decimal fractions (0,00714) for various computational purposes.
Ratios are relationships between different entities in the numerator
(a) and the denominator (b): a / b . In our case: 25/34,775. The entity in the
numerator is not reproduced in the denominator. For example, compar-
ing rates of events (incidence of events in one community) to other rates
of events (incidence of events in another community) yields ratios such as
cancer incidence rate in smokers/cancer incidence rate in nonsmokers (risk
ratio in this case). Odds (cross product ratio, relative odds) are ratios of
the probability of occurrence of an event to that of nonoccurrence of that
same event in another set of observations. In other words, it is the ratio of
the probability that something is one way to the probability that it is other
way. 64 Odds , in fact, are fractions having an event in the nominator and non-
events in the denominator: There are 100 cases of cancer and 900 healthy