Healthcare and Medicine Reference

In-Depth Information

subjects in a community of 1,000. Odds of having cancer are 100/900. Odds

in one group of subjects/odds in another group yield an odds ratio.

Ratios and odds give us an idea about
relative values
of what we see and

compare.

Odds ratio
(syn. cross-product ratio, relative odds) is the ratio of two

odds.
64

Likelihood means the general state of being likely or probable that

an event has already occurred would yield a specific outcome. Formally, it

differs from
probability
, which refers to the occurrence of future events.

Statistically speaking, a likelihood ratio is the ratio of the values of the

likelihood function at two different parameter values or under two different

data models.
64
In the domain of diagnostic methods and tests evaluation, it

is the probability that a given test result would occur in a person with the

target disorder divided by the probability that the same result would occur

in a person without that disorder.
64

Figure 2.13 represents a basic model of comparison of two groups of

events from which an event frequency is derived. It serves with other con-

siderations to assess causality, what it represents, and to what it applies

(target population or patients as individuals). A subset (sample) of indi-

viduals representing a larger (target) population to which the results of the

study should apply and be representative of is selected. Groups exposed

and nonexposed to the event (noxious factor, treatment, or other beneficial

intervention) are formed. They are compared for their “initial state” or state

before the event of interest. Then, the “intervention,” “event,” “maneuvre,”

“exposures” as they are called across the literature are compared to their

alternatives: nonevent, alternative event. Additional events (like co-morbidity

or co-treatments) may be considered. “Subsequent states,” event outcomes,

in groups are then compared for their differences and distinctions. Causality

and the role of the actions under study are clarified. Findings are then evalu-

ated based on what they actually apply to: The original target population,

individuals beyond the original target population (further generalizations),

particular and specific patients or individuals in the community, and so on.

To study and explain the cases, we need
disease frequency
, a total of

events as they were observed, and its expression in
fractions
explaining

relationships between two entities (series of observations). Fractions appear

in two forms, either as
rates
(disease or event frequency in the numerator,

population in which they occur, cases and noncases, events and nonevents

included in the denominator) or
ratios
(events in the numerator, nonevents

or other events in the denominator).

Search Pocayo ::

Custom Search