Table 3.
Hazard ratios and 95% confidence intervals of Cox proportional-hazards regression analysis during the follow-up of the study cohort.
| Characteristics | Crude HR (95% CI) |
p-value | Adjusted HRa (95% CI) |
p-value | Adjusted HRb (95% CI) |
p-value |
|---|---|---|---|---|---|---|
| TB | ||||||
| Yes | 2.16 (1.82–2.56) | <.001 | 2.16 (1.83–2.56) | <.001 | 9.94 (4.51–21.89) | <.001 |
| No | Ref. | Ref. | Ref. | |||
| Age group (years) | ||||||
| 20–40 | Ref. | Ref. | Ref. | |||
| 40–60 | 1.22 (.80–1.85) | .358 | 1.14 (.75–1.74) | .528 | 1.91 (1.03–3.54) | .039 |
| ≥60 | 3.41 (2.29–5.09) | <.001 | 3.03 (2.02–4.53) | <.001 | 5.57 (3.06–1.15) | <.001 |
| TB and age group | ||||||
| With TB and age in 40–60 | .28 (.12–.67) | .004 | – | .29 (.12–.68) | .005 | |
| Without TB and age in 40–60 | Ref. | – | Ref. | |||
| With TB and age ≥60 | .18 (.08–.41) | <.001 | – | .18 (.08–.41) | <.001 | |
| Without TB and age ≥60 | Ref. | – | Ref. | |||
| Gender | ||||||
| Male | 1.44 (1.22–1.70) | <.001 | 1.56 (1.33–1.85) | <.001 | 1.57 (1.33–1.85) | <.001 |
| Female | Ref. | Ref. | Ref. | |||
| Comorbidity | ||||||
| HTN | 1.76 (1.52–2.03) | <.001 | 1.29 (1.09–1.52) | .002 | 1.29 (1.10–1.52) | .002 |
| CAD | 1.64 (1.34–2.01) | <.001 | 1.10 (.88–1.38) | .410 | 1.10 (.87–1.38) | .422 |
| COPD | 2.06 (1.50–2.83) | <.001 | 1.15 (.83–1.60) | .409 | 1.16 (.83–1.61) | .382 |
| HF | 2.11 (1.55–2.88) | <.001 | 1.41 (1.02–1.94) | .040 | 1.39 (1.00–1.92) | .047 |
| CVD | 1.74 (1.40–2.18) | <.001 | 1.13 (.90–1.43) | .299 | 1.14 (.90–1.43) | .285 |
| LR testc, p-value | <.001 | |||||
CAD: coronary artery disease; CI: confidence interval; COPD: chronic obstructive pulmonary disease; CVD: cerebrovascular disease; HF: heart failure; HR: hazard ratio; HTN: hypertension; LR test: likelihood ratio test; T2DM: type 2 diabetes mellitus; TB: tuberculosis.
aAdjusted for age group, sex, and comorbidity.
bAdjusted for sex and comorbidity and included interaction term with age group.
cLikelihood ratio test for the full model, included interaction term of one characteristic and adjusted for others, and the nested model without interaction term.