Skip to main content

Table 2 A hypothetical population without bias

From: Detecting and correcting the bias of unmeasured factors using perturbation analysis: a data-mining approach

Level of U (i)

Population number (m i )

U is associated with neither E nor D

U is not associated with E

U is not associated with D

Exposure prevalence (p i )

Disease risk among the unexposed ( r i u )

Relative risk ( r i e / r i u )

Exposure prevalence (p i )

Disease risk among the unexposed ( r i u )

Relative risk ( r i e / r i u )

Exposure prevalence (p i )

Disease risk among the unexposed ( r i u )

Relative risk ( r i e / r i u )

1

3,000

0.40

0.30

1.50

0.40

0.40

1.54

0.80

0.30

1.50

2

2,500

0.40

0.30

1.50

0.40

0.30

1.52

0.60

0.30

1.50

3

2,500

0.40

0.30

1.50

0.40

0.25

1.45

0.40

0.30

1.50

4

2,000

0.40

0.30

1.50

0.40

0.20

1.42

0.20

0.30

1.50

Total

10,000

Crude relative risk (crude RR) = 1.50

Crude relative risk (crude RR) = 1.50

Crude relative risk (crude RR) = 1.50

Standardized relative risk (SRR) = 1.50

Standardized relative risk (SRR) = 1.50

Standardized relative risk (SRR) = 1.50