4 INITIAL STEPS
Periods
Induction LatencyDiseaseFigure 1.1 Conceptual model of the disease process.and the patient did not start smoking until September 2008, smoking could not count
as a risk factor for that patient since the induction period was too short.
Following detection, there is usually a treatment period. Following successful
treatment, the patient may become disease-free (acute illness), continue with the
illness but with fewer symptoms or signs (chronic illness), or the disease could go
into remission.
1.2.3 Estimating the Number of Persons with the Risk Factor
or Disease
In addition to modeling the course of the disease process, it is useful in planning a
study to estimate the proportion of possible subjects who have been exposed to the
risk or causal factor. For example, if the risk factor is smoking, we know that a sizable
proportion of the adult population currently smokes or has smoked in the past.
Finally, it helps to know the proportion of persons who already have a disease
in the population under study-often called the prevalence of the disease. For rare
diseases, even finding a sufficient number of cases to study can be difficult.
Sometimes in planning a study only newly diagnosed patients are used. In these
studies, we need to estimate the number of new cases that will occur in the time
interval allocated for patient entry into the study. This estimate is made from the
incidence rate for the disease being studied. The incidence rate is the number of
new cases occurring in a fixed time interval (usually 1 year) divided by the sum of
the length of time each person in a population is at risk in the interval (see Rothman
[1986] for a more complete definition). If one waits for newly diagnosed patients who
can be treated at a clinic, it is the incidence rate that is critical. For chronic diseases,
it is possible to have a rather high prevalence without having a high incidence rate if
the disease is not fatal.
In Figure 1.2, the results for five hypothetical adults that have been followed for
1 year are displayed. In this figure, D denotes a person being diagnosed with the
disease. The vertical line indicates the time t when the prevalence study occurred.
The last column gives the number of years at risk of the disease for each person. The
sum of the last column is used as the denominator in computing the incidence rate.