Study Design Classification
Comparison data required
All studies reviewed have some type of comparison group. Uncontrolled studies
will not be part of these critical reviews, except for mention in the introduction/background
section. The comparison could be to another group, whether it be part of an RCT,
a controlled trial, a case-control study, a cohort study or a comparison across
communities, or it could be over time as in a time series analysis or even a before-after
Randomized controlled trials
Study subjects (individuals or communities or groups) are randomly assigned
to 2 different groups (usually intervention and control, but may be 2 or more
interventions) and outcomes are compared. Solomon Four-Group Design and Post-Test
only Group design with randomization, and experimental studies are variations
of RCTs. Clinical trials and community trials can be RCT’s.
Controlled trials that were not randomized
A common example would be a trial within a medical practice in which the practitioner
gave the intervention to all families seen in April and May and used all families
seen in June and July as the controls. Or a surgical trial in which patients
given the new treatment are compared with a previous series of cases given the
old treatment. Or surgeon A used the new technique and compared results with
surgeon B, who used the old method. Although the assignment to one trial arm
or another is not random, the assignment is controlled by the investigator(s).
The intervention is applied to people one-at-a-time, and so the unit of the
analysis is the individual. Some quasi-experimental designs would be included
in this category. Clinical trials and community trials can be in this category.
In a cohort study, members of a defined population receive different exposures
and their outcomes are compared. Assignment of exposure is not in the hands
of the investigator. It may be decided by the individual (to smoke or not to
smoke) or by other circumstances. In a cohort study, we have at least some information
about people at the level of the individual. These are a type of observational
study; also known as longitudinal study design. Cohort studies can be prospective
Cases are persons with the outcome of interest, and controls are drawn from
the same population that produced the cases. The exposure of both groups is
then compared. This is also a type of observational study. Case-control studies
can be prospective or retrospective.
The intervention of interest is applied across an entire population or group.
The investigator has no control over the intervention. Outcomes are then measured
in the group(s) which received the intervention and the group(s) which did not
receive the intervention. The unit of analysis is the group. These studies generally
fall into 2 categories, which can be combined in a single study:
A. Ecological time study
The following study types are not included in these reviews:
The first category involves one group and the analysis measures outcomes before
and after the intervention is applied: one could call this, simply, a before-and-after
study or a time-trend study.
B. Ecological group study
The second category involves comparing two or more groups concurrently.
C. Ecological mixed study
Both within-group changes are examined and between-group changes are examined.
For example, we might compare traffic-crash mortality before and after passage
of a seat-belt law in State A, and also compare this data to mortality in
State B, which has no such law during the entire period being examined. Ecological
studies may be called many different things by different authors and perspectives:
time series studies, before and after studies, Pretest-Posttest studies, and
Only cases are described. There is no control group. For example, the authors
might report that 80% of all persons who died while riding a snow-mobile were
intoxicated with alcohol. These studies are not included in our evaluations.
This is a study which performed some type of test in something other than
a human population. For example, the ability of various life-vests to keep
a dummy afloat might be tested. These studies can be used in background section
of the review.