Table 3.
Incidence Rate Ratios (IRR) and 95% confidence intervals of CHD According to Continuous Measures of Total HDL-C (per 0.60 mmol/L), HDL-C Without ApoC-III (per 0.53 mmol/L), and HDL-C with ApoC-III (per 0.07 mmol/L) in the Nurses’ Health Study (NHS) and the Health Professionals Follow-Up Study (HPFS)
NHS | HPFS | Together | |||||
---|---|---|---|---|---|---|---|
Per SD | P | Per SD | P | P het | Per SD | P | |
Total HDL-C | |||||||
Unadjusted | 0.69 (0.54–0.87) | 0.002 | 0.70 (0.52–0.95) | 0.02 | 0.9 | 0.69 (0.58–0.83) | 0.0001 |
Adjusted | 0.77 (0.59–1.01) | 0.06 | 0.79 (0.57–1.10) | 0.2 | 0.9 | 0.78 (0.63–0.96) | 0.02 |
+ triglycerides and apoB | 0.81 (0.61–1.07) | 0.14 | 0.85 (0.61–1.18) | 0.3 | 0.8 | 0.83 (0.68–1.01) | 0.06 |
+ diabetes | 0.80 (0.60–1.05) | 0.11 | 0.86 (0.61–1.20) | 0.4 | 0.7 | 0.82 (0.66–1.02) | 0.07 |
HDL without ApoC-III | |||||||
Unadjusted | 0.55 (0.42–0.73) | 0.0001 | 0.61 (0.44–0.85) | 0.003 | 0.6 | 0.58 (0.47–0.71) | <0.0001 |
Adjusted | 0.64 (0.47–0.87) | 0.004 | 0.70 (0.50–0.97) | 0.03 | 0.7 | 0.66 (0.53–0.83) | 0.0001 |
+ triglycerides and apoB | 0.70 (0.50–0.98) | 0.04 | 0.85 (0.60–1.21) | 0.4 | 0.4 | 0.77 (0.60–0.98) | 0.03 |
+ diabetes | 0.73 (0.82–1.02) | 0.07 | 0.87 (0.61–1.25) | 0.4 | 0.5 | 0.79 (0.62–1.01) | 0.06 |
HDL with ApoC-III | |||||||
Unadjusted | 1.16 (1.00–1.35) | 0.05 | 1.20 (0.97–1.47) | 0.09 | 0.8 | 1.17 (1.04–1.32) | 0.01 |
Adjusted | 1.16 (0.99–1.35) | 0.07 | 1.21 (0.98–1.51) | 0.08 | 0.7 | 1.18 (1.03–1.34) | 0.01 |
+ triglycerides and apoB | 1.10 (0.93–1.30) | 0.3 | 1.00 (0.78–1.29) | 0.9 | 0.6 | 1.07 (0.93–1.23) | 0.4 |
+ diabetes | 1.04 (0.87–1.24) | 0.6 | 0.97 (0.76–1.25) | 0.8 | 0.7 | 1.02 (0.88–1.18) | 0.8 |
Incidence rate ratios (IRR) obtained from conditional logistic regression models.Unadjusted model takes into account age and smoking (due to matching). Multivariate model includes: alcohol, body mass index, self-reported diagnosis of hypertension before blood draw, and postmenopausal status and hormones in NHS only. HDL with and without apoC-III are simultaneously included in all models. The NHS and HPFS data were combined using random effects meta-analyses. P het=P for test of between study heterogeneity