There was no consistent effect modification by age, gender, race, or any other secondary diagnosis evaluated.
Effect modification by age was further suggested by analyses using narrower age categories.
We tested this effect modification by age formally, using likelihood ratio tests.
To assess for effect modification by age or ethnicity, stratified analyses were performed with no change in the conclusions.
A monotonic effect modification by age on the effect of the ban is further suggested by analyses based on narrower age categories (Figure 2).
Effect modification by age was investigated, with the assumption that exposure to passive smoking in public places would be greater among young people.
However, there were differences in the association between hospitalization and HRV detection when stratified by age group, evidence for effect modification by age.
We compared HRV detection among hospitalized patients and controls stratified into two groups, adult and children, due to possible effect modification by age (Table 5).
Effect modification by age was assessed by likelihood ratio tests.
No effect modification by age group was observed.
Effect modification by age will be specifically sought.
