11:709:493 Lecture Notes - Lecture 1: Randomized Controlled Trial, Observational Study, Ecological Study
This preview shows pages 1-2. to view the full 7 pages of the document.
Epidemiologic Study Design
- Can be classified as observational or experimental
▪ In observational studies, the investigator DOES NOT
assign interventions/exposures, but rather observes
subjects and measures of variables of interest
• inferences are limited to descriptions and
associations, NOT causation
▪ In experimental studies, the investigator ACTIVELY
ASSIGNS individuals/groups to
interventions/exposures under study
• It is possible to make inferences about
▪ Cohort (prospective or retrospective)
▪ Randomized controlled (clinical) trial
▪ Non-randomized trials
o Natural experiments
- Hierarchy of Scientific Evidence (higher up = better evidence)
Only pages 1-2 are available for preview. Some parts have been intentionally blurred.
o Meta-analysis of RCTs
o Systematic review
o Randomized control trial
o Non-randomized studies
o Cohort studies
o Case control studies
o Cross sectional studies
o Ecological studies
- Why design is best?
o depends on the study question
▪ what is the current knowledge on the topic?
▪ How common is the disease/risk factors?
▪ How long would the study take?
▪ What is the budget?
▪ Are their ethical issues involved?
Cross Sectional Studies
- Observational study conducted at a defined point in time with
one measurement per unit of analysis
- Allows you to estimate the prevalence of a disease
o Group 1, group 2, group 3 = compared at one time
o Short duration
o Easy to implement
You're Reading a Preview
Unlock to view full version