Effect estimates for association between sub-optimal LDL-C response at 24 months to initiated statin therapy and the risk of incident cardiovascular events using different statistical approaches (n=165 411)
| Group | Number of CVD events | Rate of CVD events (per 1000 person-years) | Crude/unadjusted models | Adjusted models | |||
| Cox regression | Competing-risks survival regression | Cox regression | Competing-risks survival regression | ||||
| HR (95% CI) | sHR (95% CI) | HR (95% CI) | sHR (95% CI) | ||||
| Overall CVD-related event* | Optimal | 10 656 | 19.7 | 1 | 1 | 1 | 1 |
| Sub-optimal | 12 142 | 22.6 | 1.17 (1.13 to 1.20) | 1.13 (1.10 to 1.16) | 1.22 (1.19 to 1.25) | 1.19 (1.16 to 1.23)† | |
| CAD‡ | Optimal | 5986 | 11.1 | 1 | 1 | 1 | 1 |
| Sub-optimal | 7156 | 13.3 | 1.22 (1.17 to 1.26) | 1.18 (1.14 to 1.22) | 1.30 (1.25 to 1.34) | 1.23 (1.19 to 1.27) § | |
| Stroke/TIA* | Optimal | 2387 | 4.4 | 1 | 1 | 1 | 1 |
| Sub-optimal | 2478 | 4.6 | 1.07 (1.01 to 1.13) | 1.02 (0.96 to 1.08) | 1.15 (1.08 to 1.22) | 1.10 (1.04 to 1.17)¶ | |
| PVD* | Optimal | 1462 | 2.7 | 1 | 1 | 1 | 1 |
| Sub-optimal | 1635 | 3.0 | 1.14 (1.07 to 1.23) | 1.09 (1.02 to 1.17) | 1.16 (1.08 to 1.25) | 1.12 (1.04 to 1.21)** | |
| CVD-related death* | Optimal | 821 | 1.5 | 1 | 1 | 1 | 1 |
| Sub-optimal | 873 | 1.6 | 1.09 (0.99 to 1.20) | 1.04 (0.95 to 1.15) | 1.25 (1.13 to 1.38) | 1.21 (1.09 to 1.34)†† | |
Cox regression provides hazard ratio (HR) whereas competing-risks survival regression (Fine-Grey model) provides sub-hazard ratio (sHR).
*The multivariable Cox and competing-risk regression models for overall CVD-related events, stroke/TIA, PVD and CVD-related deaths were adjusted for age and baseline LDL-C value.
†Competing risks for overall CVD-related event were non-CVD-related death and transfer out of practice.
‡The multivariable Cox and competing-risk regression models for CAD were adjusted for age.
§Competing risks for CAD model were death, transfer out of practice, stroke/TIA and PVD.
¶Competing risks for stroke/TIA model were death, transfer out of practice, CAD and PVD.
**Competing risks for PVD model were death, transfer out of practice, CAD and stroke/TIA.
††Competing risks for CVD-related death were non-CVD-related death, transfer out of practice, CAD, stroke/TIA and PVD.
CAD, coronary artery disease; CPRD, Clinical Practice Research Datalink; CVD, cardiovascular disease; HES, Hospital Episodes Statistics, LDL-C, low-density lipoprotein cholesterol; ONS, Office of National Statistics; PVD, peripheral vascular disease; TIA, transient ischaemic attack.