Observational Studies: Descriptive & Analytical Types
Observational studies involve no intervention other than asking questions and carrying
out medical examinations and simple laboratory tests or X-ray examinations. In
epidemiology, observational studies are more common than experimental ones,
particularly if an investigator wants to determine whether an agent or exposure causes
cancer in humans.
Descriptive studies tend to be simpler and easier to conduct than analytical or
experimental studies but they are nonetheless quite important. Descriptive studies can
provide the background from which analytical studies emerge. They help to generate
hypotheses, as opposed to testing them.
- A large range of outcomes because no subgroups are studied
- A large range of potential predictors again because no subgroups are studied
- Not possible to study subgroups
- No control for confounding as data is in aggregate form
- Not able to reproduce/replicate results as data was not collected in an experiment
with defined perimeters.
1. Cross-Sectional Comparison Studies: "Am I like my neighbors?"
Cross-sectional studies compare data that are combined from smaller groups as
opposed to very large descriptive studies. These studies focus on observations made at
only one point in time so they are quickly completed and relatively inexpensive. But
they cannot reveal a sequence of events over time since they sample data only once.
Cross-sectional studies often simply compare the rate of a particular cancer in one
place versus another place.
2. Correlation (Ecologic) Studies: "What if I am exposed to this?"
Ecologic studies look at diet and cancer at the population level, think of this as the view
from 30,000 feet. These types of studies represent a transition to analytical studies
since they compare cancer rates of populations in relation to risk factors. They do not
include outcome so they aren't considered analytical.
- The diet-cancer correlation
- Following populations as they migrate to compare cancer rates
Ecologic studies can provide powerful clues pointing in a particular direction, especially
when they compare large populations with different diets.