Table 4.
Variable | Total N | Sepsis events, | Sepsis incidence (per 1000 person-years), | Models for sepsis hazard | |
---|---|---|---|---|---|
Unadjusted, | Adjusteda, | ||||
N (%) | IR (95% CI) | HR (95% CI) | HR (95% CI) | ||
Statin non-users | |||||
HDL-C quartile | |||||
Q1 (5–41 mg/dl) | 5233 | 315 (6.0) | 9.6 (8.6–10.7) | 1.73 (1.43–2.09) | 1.09 (0.88–1.36) |
Q2 (42–50 mg/dl) | 4540 | 223 (4.9) | 7.9 (6.9–9.0) | 1.43 (1.16–1.76) | 1.11 (0.89–1.37) |
Q3 (51–62 mg/dl) | 4751 | 201 (4.2) | 6.8 (5.9–7.8) | 1.21 (0.98–1.49) | 1.06 (0.86–1.31) |
Q4 (63–199 mg/dl) | 4837 | 169 (3.5) | 5.5 (4.8–6.4) | Reference | Reference |
Trend p value | <0.001 | 0.44 | |||
LDL-C quartile | |||||
Q1 (3–99 mg/dl) | 4816 | 278 (5.8) | 9.4 (8.3–10.5) | 1.54 (1.28–1.86) | 1.32 (1.10–1.60) |
Q2 (100–120 mg/dl) | 4810 | 231 (4.8) | 7.6 (6.7–8.7) | 1.29 (1.06–1.57) | 1.19 (0.98–1.46) |
Q3 (121–143 mg/dl) | 4842 | 212 (4.4) | 6.9 (6.0–7.9) | 1.13 (0.93–1.37) | 1.11 (0.91–1.36) |
Q4 (144–388 mg/dl) | 4619 | 169 (3.7) | 5.8 (5.0–6.8) | Reference | Reference |
Trend p value | <0.001 | 0.003 | |||
Statin users | |||||
HDL-C quartile | |||||
Q1 (11–39 mg/dl) | 2304 | 170 (7.4) | 11.8 (10.2–13.7) | 1.37 (1.08–1.73) | 0.89 (0.69–1.17) |
Q2 (40–47 mg/dl) | 2274 | 134 (5.9) | 9.6 (8.1–11.4) | 1.11 (0.87–1.43) | 0.87 (0.67–1.13) |
Q3 (48–57 mg/dl) | 2163 | 119 (5.5) | 8.8 (7.3–10.5) | 1.03 (0.80–1.33) | 0.93 (0.72–1.20) |
Q4 (58–156 mg/dl) | 2178 | 119 (5.5) | 8.7 (7.3–10.4) | Reference | Reference |
Trend p value | 0.006 | 0.39 | |||
LDL-C quartile | |||||
Q1 (11–76 mg/dl) | 2210 | 175 (7.9) | 13.1 (11.3–15.1) | 2.00 (1.55–2.59) | 1.50 (1.14–1.96) |
Q2 (77–93 mg/dl) | 2304 | 142 (6.2) | 9.9 (8.4–11.7) | 1.55 (1.19–2.03) | 1.36 (1.03–1.78) |
Q3 (94–111 mg/dl) | 2091 | 127 (6.1) | 9.6 (8.1–11.4) | 1.49 (1.13–1.96) | 1.38 (1.04–1.81) |
Q4 (112–315 mg/dl) | 2191 | 81 (3.7) | 5.9 (4.7–7.3) | Reference | Reference |
Trend p value | <0.001 | 0.008 |
Total of 29,690 REGARDS participants. HRs estimated using Cox proportional hazards regression and represent pooled estimates from multiple imputation
p values represent tests for linear trends across quartiles
aAdjusted for demographics, health behaviors, chronic medical conditions, biomarkers, and functional status measures. Demographics = age, gender, race, region, income, and education; health behaviors = smoking status and alcohol use; chronic medical conditions = cancer, chronic lung disease, coronary artery disease, diabetes, hypertension, obesity, stroke, and Morisky medication adherence index; biomarkers = hs-CRP, Cystatin C, ACR, and eGFR; functional status measures = weakness, exhaustion, and low physical activity
HR hazard ratio, IR incidence rate, CI confidence interval, LDL-C low-density lipoprotein, HDL-C high-density lipoprotein, hs-CRP high-sensitivity C-reactive protein, ACR albumin–creatinine ratio, eGFR estimated glomerular filtration rate