Table 3.
Association between baseline TyG and incident heart failure in the Kailuan cohort
| Q1 | Q2 | Q3 | Q4 | P for trend | Per 1-unit increment | P value | |
|---|---|---|---|---|---|---|---|
| Number of patients | 23,997 | 24,000 | 24,001 | 23,998 | |||
| HF cases | 497 | 576 | 729 | 924 | |||
| Persons | 23,997 | 24,000 | 24,001 | 23,998 | |||
| Person-years | 297,098 | 295,908 | 294,857 | 292,552 | |||
| HF incidence 1 | 1.67(1.53–1.83) | 1.95(1.79–2.11) | 2.47(2.30–2.66) | 3.16(2.96–3.37) | |||
| Model 1 2 | 1 | 1.16(1.03–1.31) | 1.48(1.32–1.66) | 1.90(1.70–2.11) | < 0.001 | 1.44(1.37–1.51) | < 0.001 |
| Model 2 3 | 1 | 1.12(1.00-1.27) | 1.41(1.26–1.58) | 1.90(1.70–2.12) | < 0.001 | 1.47(1.40–1.55) | < 0.001 |
| Model 3 4 | 1 | 1.00(0.88–1.12) | 1.12(1.00-1.26) | 1.23(1.09–1.39) | < 0.001 | 1.17(1.10–1.24) | < 0.001 |
| Sensitivity analysis | |||||||
| Sensitivity analysis 1 5 | 1 | 1.00(0.88–1.13) | 1.14(1.01–1.28) | 1.23(1.10–1.41) | < 0.001 | 1.18(1.10–1.25) | < 0.001 |
| Sensitivity analysis 2 6 | 1 | 0.97(0.84–1.11) | 1.11(0.98–1.27) | 1.17(1.02–1.34) | 0.006 | 1.13(1.06–1.22) | < 0.001 |
| Competing risk regression 7 | 1 | 1.01(0.90–1.14) | 1.14(1.02–1.29) | 1.25(1.10–1.41) | < 0.001 | 1.18(1.10–1.25) | < 0.001 |
1 HF incidence; The incidence rates per 1000 person years with the corresponding 95% confidence intervals are shown
2 Model 1: Unadjusted. The hazard ratios with the corresponding 95% confidence intervals are shown
3 Model 2: Age-sex adjusted. The hazard ratios with the corresponding 95% confidence intervals are shown
4 Model 3: Adjusted for age, gender, education, income, physical activity, smoking status, alcohol intake, diabetes, LDL-c, HDL-c, SBP, DBP, BMI, eGFR, hs-CRP, anti-hypertensive drugs, anti-diabetes drugs, and lipid-lowering drugs. The hazard ratios with the corresponding 95% confidence intervals are shown
5 Sensitivity analysis 1: exclude follow-up time less than 2 years, remained 95,275 subjects with 2,497 HF cases. The hazard ratios with the corresponding 95% confidence intervals are shown
6 Sensitivity analysis 2: exclude subjects with medication at baseline (anti-hypertension drugs, lipid lower drugs, anti-diabetes drugs), remained 85,118 subjects with 2,118 HF cases. The hazard ratios with the corresponding 95% confidence intervals are shown
7 Competing risk regression: sub-hazard ratios with the corresponding 95% confidence intervals are shown