Table 2.
Incidence rate difference and hazard ratio for primary and secondary endpoints under different analysis model
| Outcome | Unmatched | PSM (1:1) | IPTW | ||||||
|---|---|---|---|---|---|---|---|---|---|
| IRD (100 person-month; 95% CI)a | Adjusted HR (95% CI)b | P-value | IRD (100 person-month; 95% CI)a | Adjusted HR (95% CI)c | P-value | IRD (100 person-month; 95% CI)a | Adjusted HR (95% CI)d | P-value | |
| Primary endpointe | −0.39(−0.54 to −0.25) | 0.62(0.43–0.89) | 0.01 | −0.32(−0.52 to −0.11) | 0.55(0.37–0.82) | 0.003 | −0.31(−0.46 to −0.15) | 0.66(0.44–0.98) | 0.04 |
| Second endpoints | |||||||||
| All-cause rehospitalization | −0.34(−0.63 to −0.04) | 0.92(0.72–1.16) | 0.47 | −0.20(−0.58 to 0.19) | 0.86(0.67–1.10) | 0.23 | −0.34(−0.64 to −0.04) | 0.94(0.75–1.18) | 0.58 |
| Cardiovascular rehospitalization | −0.16(−0.41 to 0.10) | 1.02(0.78–1.33) | 0.91 | −0.11(−0.45 to 0.22) | 0.88(0.67–1.16) | 0.38 | −0.19(−0.46 to 0.07) | 1.00(0.78–1.28) | 0.99 |
| Heart failure rehospitalization | −0.31(−0.45 to −0.18) | 0.61(0.41–0.92) | 0.02 | −0.29(−0.48 to −0.10) | 0.52(0.34–0.80) | 0.003 | −0.27(−0.40 to −0.13) | 0.62(0.41–0.95) | 0.03 |
| All-cause death | −0.15(−0.22 to −0.08) | 0.53(0.25–1.14) | 0.11 | −0.07(−0.17 to 0.02) | 0.57(0.26–1.27) | 0.17 | −0.09(−0.17 to −0.02) | 0.61(0.26–1.38) | 0.23 |
| Cardiovascular death | −0.09(−0.14 to −0.04) | 0.47(0.15–1.44) | 0.19 | −0.04(−0.10 to 0.03) | 0.54(0.17–1.77) | 0.31 | −0.05(−0.10 to 0.01) | 0.63(0.18–2.22) | 0.47 |
| Stroke | −0.04(−0.10 to 0.02) | 0.60(0.26–1.40) | 0.24 | −0.03(−0.11 to 0.05) | 0.69(0.28–1.66) | 0.40 | −0.05(−0.11 to 0) | 0.58(0.26–1.32) | 0.20 |
CI, confidence interval; HR, hazard ratio; IPTW, inverse probability of treatment weighting; IRD, incidence rate differences; PSM, propensity score matching.
aIRD indicates an absolute difference in the incidence rate between the two groups.
bHRs estimated using a Cox proportional hazard model and adjusted for potential confounders (unbalanced baseline variables), including age, body mass index, CHA2DS2-VASc score, stroke, chronic kidney disease, chronic obstructive pulmonary disease, N-terminal pro-brain natriuretic peptide, serum creatinine, left atrial diameter, diastolic blood pressure, heart rate, aldosterone antagonist, diuretic, digoxin, and oral anticoagulant in all HFpEF patients for the primary endpoint and second endpoints.
cHRs estimated using a Cox proportional hazard model in patients after matching.
dHRs estimated using a Cox proportional hazard model in all patients with assigning each patient a weight on the basis of the propensity score.
eThe primary endpoint is a composite of all-cause death or rehospitalization for worsening heart failure.