Table 2.
Persistent exposure required * |
Insulin only
N = 1596 |
Sulfonylurea + Insulin
N = 1596 |
---|---|---|
Composite fatal and nonfatal cardiovascular events | 128 | 134 |
Person-years | 2597 | 2394 |
Unadjusted rate/1000 person-years (95 % CI) | 49.3 (41.2, 58.3) | 56.0 (47.1, 65.9) |
Hazard ratio† (95 % CI) | 0.89 (0.70, 1.10) | Reference |
First hypoglycemia events | 179 | 174 |
Person-years | 2420 | 2206 |
Unadjusted rate/1000 person-years | 74.0 (63.9, 85.6) | 78.9 (68.0, 91.5) |
Hazard ratio† (95 % CI) | 0.94 (0.76, 1.16) | Reference |
Recurrent hypoglycemia events | 267 | 287 |
Person-years | 2669 | 2458 |
Unadjusted rate/1000 person-years | 100.0 (85.1, 118.3) | 116.8 (98.5, 139.4) |
Hazard ratio† (95 % CI) | 0.87 (0.69, 1.10) | Reference |
Composite chronic kidney disease events | 218 | 174 |
Person years | 2296.3 | 2216.4 |
Unadjusted rate/1000 person-years | 94.9 (83.3, 107.7) | 78.5 (67.7, 90.5) |
Hazard ratio† (95 % CI) | 1.22 (1.00, 1.49) | Reference |
All-cause death | 285 | 267 |
Person years | 2667 | 2453 |
Unadjusted rate/1000 person-years | 106.9 (95.4, 119.2) | 108.8 (96.8, 121.8) |
Hazard ratio† (95 % CI) | 0.99 (0.84, 1.17) | Reference |
*Primary analysis requires persistence on insulin; patients are censored after 90 days without any insulin available. Patients were followed until each individual outcome and other events were ignored, therefore the amount of person time varies in each analysis
†Propensity-score matched hazard is derived from Cox proportional hazards model for time to outcome