Table 4.
Rosiglitazone use | Setting entry date to 1 January 2005 | Deleting patients who developed skin cancer within 3 months of follow-up | Excluding patients with a history of any cancer before 2006 | Including pioglitazone users in the analyses | Time-dependent models | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
HR | 95% CI | P | HR | 95% CI | P | HR | 95% CI | P | HR | 95% CI | P | HR | 95% CI | P | |
Ever users | 0.986 | (0.882, 1.102) | 0.7977 | 0.933 | (0.809, 1.075) | 0.3368 | 0.951 | (0.834, 1.083) | 0.4469 | 0.938 | (0.835, 1.055) | 0.2863 | 0.909 | (0.785, 1.051) | 0.1981 |
Duration of therapy (months) | |||||||||||||||
<3.73 | 1.177 | (0.975, 1.421) | 0.0903 | 1.119 | (0.900, 1.392) | 0.3106 | 1.055 | (0.855, 1.302) | 0.6199 | 1.114 | (0.930, 1.336) | 0.2411 | 1.287 | (1.050, 1.579) | 0.0153 |
3.73-13.77 | 1.112 | (0.939, 1.318) | 0.2189 | 0.989 | (0.792, 1.234) | 0.9194 | 1.096 | (0.906, 1.325) | 0.3452 | 1.025 | (0.862, 1.219) | 0.7781 | 1.059 | (0.851, 1.317) | 0.6099 |
>13.77 | 0.787 | (0.662, 0.935) | 0.0065 | 0.713 | (0.554, 0.918) | 0.0087 | 0.737 | (0.592, 0.916) | 0.0060 | 0.718 | (0.589, 0.874) | 0.0010 | 0.583 | (0.457, 0.740) | <0.0001 |
P-trend | 0.1215 | 0.0484 | 0.0871 | 0.0186 | 0.0024 | ||||||||||
Cumulative dose (mg) | |||||||||||||||
<448 | 1.138 | (0.939, 1.379) | 0.1883 | 1.092 | (0.876, 1.361) | 0.4336 | 0.985 | (0.793, 1.223) | 0.8882 | 1.059 | (0.880, 1.275) | 0.5418 | 1.286 | (1.030, 1.607) | 0.0264 |
448-1752 | 1.099 | (0.929, 1.301) | 0.2722 | 0.923 | (0.735, 1.159) | 0.4892 | 1.105 | (0.915, 1.334) | 0.3015 | 1.011 | (0.849, 1.203) | 0.9047 | 0.983 | (0.775, 1.247) | 0.8876 |
>1752 | 0.815 | (0.687, 0.966) | 0.0186 | 0.798 | (0.627, 1.016) | 0.0672 | 0.780 | (0.630, 0.965) | 0.0224 | 0.773 | (0.638, 0.936) | 0.0084 | 0.614 | (0.476, 0.792) | 0.0002 |
P-trend | 0.1839 | 0.0971 | 0.1685 | 0.0482 | 0.0041 |
Referent group: never-users of rosiglitazone; HR: hazard ratio, CI: confidence intervals.
In models excluding patients not treated with rosiglitazone and using the first tertile of exposure as referent, the hazard ratio (95% confidence interval) for the second and third tertile of duration of therapy was 0.850 (0.634, 1.140) (P = 0.2777) and 0.639 (0.465, 0.876) (P = 0.0054), respectively (P-trend = 0.0054); and was 0.852 (0.633, 1.148) and 0.712 (0.521, 0.973), respectively, for cumulative dose (P-trend = 0.0327).