Table 3. Risk estimates of lipid biomarkers on mortality as in the summarised reports.
Garcia-Palmieri et al. (1981) reported multiple regression coefficients (“The logistic function was used to model the relationship between cancer mortality and serum cholesterol”), so that the OR herein reported corresponds to the exponential function of the coefficients in the manuscript. Crespo et al. (2002) - The authors did not provide risk estimates; the IRR herein reported corresponds to the relation between crude death rates among people with high cholesterol (crude death rate = 16.3%), borderline high blood cholesterol (crude death rate = 15.5%) and desirable cholesterol (crude death rate = 15.8%). For example, 15.8/16.3 ∼0.97. Where available, the risk estimates herein summarized correspond to the adjusted models reported in the original reports.
| Author | Exposure definition | All-cause mortality | Cardiovascular mortality | Other causes of death assessed | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Risk estimate | lower IC | upper IC | Risk estimate | lower IC | upper IC | Risk estimate | lower IC | upper IC | Cause? | ||
| Lazo-Porras et al. (2016) | Normal HDL vs isolated low HDL [HDL-c <40 mg/dL in men and <50 mg/ dL in women without hypertriglyceridemia and LDL <160 mg/dL] | RR = 1.11 | 0.49 | 2.51 | RR = 0.45 | 0.05 | 4.24 | ||||
| Normal HDL vs non-isolated low HDL [HDL-c <40 mg/dL in men and <50 mg/dL in women with hypertriglyceridemia and/or LDL ≥160 mg/dL] | RR = 0.82 | 0.21 | 3.15 | RR = 3.67 | 0.48 | 28.16 | |||||
| Werle et al. (2011) | LDL (mg/dL) | HR = 1.00 | 0.99 | 1.01 | |||||||
| HDL (mg/dL) | HR = 1.01 | 0.97 | 1.05 | ||||||||
| ApoA-I (mg/dL) | HR = 0.99 | 0.98 | 1.01 | ||||||||
| Garcia-Palmieri et al. (1981) | Serum cholesterol (mg/dL), 45–54 rural men | OR = 0.54 | p < 0.05 | Cancer | |||||||
| Serum cholesterol (mg/dL), 55–64 rural men | OR = 0.48 | p < 0.05 | Cancer | ||||||||
| Serum cholesterol (mg/dL), 45–54 urban men | OR = 0.78 | Cancer | |||||||||
| Serum cholesterol (mg/dL), 55–64 urban men | OR = 1.08 | Cancer | |||||||||
| Crespo et al. (2002) | Desirable total cholesterol ( <200 mg/dL) vs high total cholesterol (≥240 mg/dL) | IRR = 0.97 | |||||||||
| Desirable total cholesterol ( <200 mg/dL) vs borderline high blood cholesterol (200-239 mg/dL) | IRR = 1.02 | ||||||||||
| Garcia-Palmieri et al. (1988) | Serum cholesterol (mg/dL) | OR = 0.004 | p < 0.05 | ||||||||