Relative Risk Estimation Using MH Method For ES Reports

The statistical methods used to describe the incidence of absences are the relative risk and the 95% confidence interval. The relative risk is the rate of absence in one group divided by the rate in a reference (comparison) group.

For the Mantel-Haenszel (MH) method* the reference group is all workers other than the occupational category of primary interest. A relative risk of 1.0 indicates that both groups have the same risk of absence. A relative risk greater than 1.0 indicates that workers in a selected occupational category have a higher risk of absence than workers in all other occupational categories combined. A relative risk less than 1.0 implies that the selected occupational group has a lower risk of absence compared to all other occupational categories combined.

An example for All Diseases and Injuries is given in Table 10 of The Abridged Rocky Flats 1993 Report

A detailed example of the MH Method is given on the Biostatistical Methods page .

*Greenland,S and J.M. Sanders (1985), "Estimation of a Common Effect Parameter from Sparse Follow-Up Data", Biometrics 41, 55-68.