Table 4

Crude and adjusted rate ratios (RRs) of death for inhaled corticosteroid use relative to bronchodilator use, from the Saskatchewan chronic obstructive pulmonary disease (COPD) cohort, comparing three approaches to data analysis and two cohort entry definitions

Using first prescription as cohort entry
 Hierarchical intention-to-treat analysis0.520.660.57–0.76
 Conventional intention-to-treat analysis0.630.750.62–0.90
 According-to-treatment analysis0.670.850.73–0.98
Using third prescription as cohort entry
 Hierarchical intent-to-treat analysis0.530.710.60–0.83
 Conventional intention-to-treat analysis0.570.720.58–0.88
 According-to-treatment analysis0.700.940.81–1.09
  • CI: confidence interval

  • #: adjusted for age at cohort entry and sex, as well as the occurrence of COPD hospitalisations and the use of other COPD medications dispensed during the 6‐month period prior to cohort entry