Ecological Studies

1
Ecological Studies

• By Rune Jacobsen

2

• Part1 - Ecological studies theoretical
• Part 2 - An example

3
The ecological study

• Comparison of groups rather than individuals
• Missing individual information
• Low cost
• Interest in ecological effects

4
Levels of measurements

• Aggregate measures
• Summaries of observations at individual level
  e.g., proportion of smokers
• Environmental measures
• Physiological characteristics
  e.g., air pollution levels
• Global measures
• GNP, population density

5
Levels of analysis

• Individual level e.g. average level of air
  pollution is assigned to each individual, and
  indidvidual age, sex, ethnicity, smoking status
  are known
• Partially ecological analysis E.g. some
  variables known for individuals (age, sex, air
  pollution) but others for the population (
  smokers, vegetarians)
• Fully ecological analysis Information on
  exposure and disease are only known for the
  population (GNP and national death rates)

6
Levels of Inference

• Biological (individual use of MC helmet)
• Ecological (groups Effect of MC helmet law)
• Contextual effect
• Example effect law ? helmet use on risk of MC-
  mortality controlling for individual helmet use

7
The ecologically fallacy
8
Ecollogical fallacy

• Across level inference. E.g. from group to
  individual.

9
Counfounding ecological study

• Example

10
Example
11
Example
12
Example with no confounding by group (but there
could be confounding at individual level)
13
Example with confounding by group
14
Example with effect modification by group
15
Problems with ecological studies

• Ecological bias, in estimating effects at the
  individual level may result from
• Within group bias if there is confounding,
  selection bias or misclassification within each
  group then the ecological estimate may also be
  biased
• Confounding by groups the background disease
  rate varies across groups
• Effect modification by group the excess rate
  due to exposure varies across groups

16
Problems with ecological studies

• The major problem of ecological bias arise from
  attempts at cross-level inference, e.g. in
  studies where the intention is to make inferences
  at the individual level
• Neverttheless ecological studies have played a
  major role in the development, and to some extent
  in the testing, of epidemiological hypothesis.

17
Illustration by example

• Part II - Test of a hypothesis
• The queen in the lancet