|
|
|
Department of Family Medicine
EBM Study Design
Descriptive
-
Case Study–Description of the characteristics,
treatments, &/or outcomes of one case
-
Case Series–Description of the
characteristics, treatments, &/or outcomes of several cases
-
Cross-Sectional Survey (Determination of the
prevalence of disease, exposures, outcomes of a given disease at a given
time in the population)
Strengths
-
Cheap and simple
-
Ethically safe
Limitations
-
Establishes association at most,
not causality
-
Recall bias susceptibility
-
Confounders may be unequally
distributed
-
Neyman bias
-
Group sizes may be unequal
Analytic
1. Experimental (randomized control trial) [RCT]
-
Investigator has control over exposure
(intervention) of study subjects and this is allocated on a random basis.
-
Blinding: Single blind/Double blind
-
Primary analysis is based upon the group to
which the subjects were assigned. Known as INTENTION TO TREAT ANALYSIS.
| Strengths |
Limitations |
-
Avoids selection bias
-
Provides clear evidence that the
intervention precedes the cause
-
Allows the determination of the incidence
of the side effects and complications of treatment
-
Most convincing study design
-
Unbiased distribution of confounders
-
Blinding more likely
-
Randomization facilitates statistical
analysis
|
|
2. Observational
A. Cohort Study
-
Comparison groups formed on the basis of
EXPOSURE (yes/no) and followed in time for the occurrence of DISEASE
(yes/no)
-
Differs from a RCT in that there is no random
assignment in a cohort study.
-
Differs from a case-control study in that groups
are formed based upon exposure and not disease.
-
Two types:
- Prospective
- Retrospective
| Strengths |
Limitations |
-
Exposure occurs before disease
-
Can measure incidence in the exposed and
unexposed groups and can calculate relative risk and risk difference
(attributable risk)
-
Can investigate multiple outcomes with a
single exposure.
-
Ethically safe
-
Subjects can be matched
-
Can establish timing and directionality of
events
-
Eligibility criteria and outcome
assessments can be standardized
-
Administratively easier and cheaper than
RCT
|
-
Exposure status is self-selected and may
be linked to a hidden confounder
-
Cost of doing this study is high
(millions)
-
Must follow groups a long time for
outcomes to occur. Not good for rare outcomes.
-
Controls may be difficult to identify
-
Blinding is difficult
-
Randomization not present
|
- Relative Risk = incidence in the exposed/incidence in the unexposed. Tells
you how many times more likely to get a disease is someone exposed compared
with someone not exposed.
- Risk Difference (Attributable Risk) = incidence in the exposed - incidence
in the unexposed. Tells you the excess amount of disease due to exposure.
Tells you what proportion of new cases could have been prevented by
treatment.
B. Case-control studies
Comparison groups formed on a basis of a DISEASE (yes/no), then look back in
time to assess EXPOSURE (yes/no).
| Strengths |
Limitations |
-
Good for rare disease
-
Relatively inexpensive (hundreds of
thousands)
-
Can investigate many exposures at the same
time
-
Ethically safe
-
Subjects can be matched
-
Can establish timing and directionality of
events
-
Eligibility criteria and outcome
assessments can be standardized
-
Administratively easier and cheaper than
RCT
|
-
Relies upon recall of past exposure
information.
-
Cannot calculate disease incidence
-
Difficult to select a control group for
comparison (controls may be difficult to identify)
-
Exposure may be linked to a hidden
confounder
-
Blinding is difficult
-
Randomization not present
-
For rare disease, large sample sizes or
long follow-up necessary
|
Analysis of the data is as follows:
|
|
Cases |
Controls |
| Exposed |
Yes |
a |
b |
|
No |
c |
d |
Odds Ratio = (ad/bc) and is a good estimate of the relative risk
|