Basic Statistics

COMMON TYPES OF BIOMEDICAL STUDIES 9

1.3.4 Field Trials

Field trials are performed using subjects who are not patients and hence often require
visiting the subjects in the field, which may be a school, church, or other location
convenient for the subjects. One example of a field trial would be a vaccine trial.
One of the largest vaccine trials was for the Salk vaccine, developed to prevent polio.
A vaccine or placebo was given to over 1,000,000 schoolchildren and the incidence
of polio in the treated and untreated groups was assessed. It was necessary to make
the sample size large because the incidence of polio was not high. Randomization
of individuals is often difficult to do in field trials because of difficulty in obtaining
acceptance in the field.

1.3.5 Prospective Studies

In a typical prospective study, no treatment is assigned. There are three general types
of prospective studies.

1. In prospective trend studies, repeated samples of different individuals can be
   taken at intervals over time from a dynamic population where some of the
   individuals in the population may change over time. Such studies have been
   used to study voting intentions as an election approaches.


2. In prospective cohort studies, repeated samples of different subjects are taken
   from the same cohort (group of people). Such studies have been used to deter-
   mine how students change their attitudes as they progress in school.


3. In prospective panel studies, repeated measures are made on the same individ-
   uals over time. This is the type of prospective study most used in biomedical
   studies. Hereafter in this book, if the term prospective study is used, it will
   be assumed that a prospective panel study is meant. Note that epidemiologists
   sometimes call panel studies cohort studies or simply prospective studies.

In the classical epidemiological prospective (panel) study, a cohort of disease-free
individuals are measured for exposure to the causal factor(s) at the beginning of the
follow-up period. Then, at subsequent examinations, exposure can be remeasured
and disease status (outcome) must be measured. In prospective panel studies, the
subjects can be classified into two groups based on their exposure status. The critical
point is that the measurement of exposure precedes the measurement of the outcome
in time. There is no opportunity for the disease to have an effect on the exposure
factor.
To use this design, it is necessary that the disease being studied have a high in-
cidence rate in order to accumulate enough cases (persons with the disease) in a
reasonable follow-up period. The disease studied must be severe enough that its
occurrence can be detected. A sufficient number of persons with exposure must be
available at the start of the study. To obtain a sufficiently high incidence rate of the
disease in the study group, high-risk groups are often followed.