Last update:
11-Nov-2004

Stratified analysis or stratification is used to study and control for the effect of other than variables exposure and disease, in data analysis. Stratification means that the data are separated into categories or strata. For example stratification by sex or age means that the data are separated into male and female groups or into categories by age. The aims of stratification in an epidemiological study are control of confounding and analysis of effect measure modification. A simple definition of confounding would be the confusion, or mixing, of effects. This definition implies that the effect of the exposure is mixed with the effect of another variable, leading to a bias. When data are stratified by confounder, for example, into men and women, each stratum would be free of confounding by that variable. Thus, if the correlation between exposure and disease is analyzed separately in men and women, each category of gender would give an estimation of the result of exposure, independent of gender. A well-designed study requires control for confounding, which can be achieved in various ways, apart from stratification, such as adjustment, standardization, restriction and experiment. Effect measure modification should be taken into account during the planning, analysis and interpretation of results of epidemiological studies. Modification exists when the relation between two variables is different depending on the levels of a third variable. For example, obesity increases the risk for breast cancer in postmenopausal but not in premenopausal women. In this way the presence or absence of menopause modifies the effect of obesity on breast cancer. Thus menopausal status is an effect modifier of obesity on breast cancer. The modifier is the characteristic of individuals which modifies the effect of an exposure on the incidence of a disease, or in other words, the characteristic that modifies the relation between exposure and disease.

Key words: Confounder, Confounding, Effect measure modification, Stratification, Stratified analysis.