Skip to main content

Table 1 Predictors of non-terminal and terminal events utilizing Bayesian Independent AFT model with log-Normal baseline survival distribution

From: A bayesian approach to model the underlying predictors of early recurrence and postoperative death in patients with colorectal Cancer

 

Recurrence

Death without recurrence

Death after recurrence

TR

95% CI

TR

95% CI

TR

95% CI

Age at Diagnosis (years)

Trend effect

0.764

0.538–0.935*

0.396

0.313–0.532 *

0.659

0.559–0803 *

Gender

Male

0.596

0.456–0.855 *

0.666

0.363–1.074

0.951

0.697–1.203

Metastasis to other sites

Yes

0.735

0.589–1.022

0.566

0.490–0.835*

0.821

0.594–1.052

Number of chemotherapies

Trend effect

1.583

1.367–1.863 *

1.541

1.332–2.243 *

2.029

1.792–2.191 *

Grade

(differentiation level)

Well

Referent

---

---

---

---

---

Moderate

0.648

0.503–0.955 *

0.527

0.387–0.674 *

1.053

0.839–1.277

Poor

0.856

0.733–0.988 *

0.558

0.380–1.030

1.310

0.875–1.174

Tumor size

Trend effect

0.709

0.610–0.929 *

0.595

0.500–0.679 *

0.893

0.781–1.010

PT-Stage

T2

Referent

---

---

---

---

---

T3

0.601

0.530–0.713 *

0.363

0.592 − 0.301 *

0.983

0.746–1.414

T4

0.962

0.872–1.319

0.434

0.347–0.545*

0.853

0.509–1.402

TX

NC

NC

NC

NC

NC

NC

PN-Stage

N0

Referent

---

---

---

---

---

N1

1.272

0.967–1.586

1.947

1.728–2.122

0.760

0.576–1.468

N2

0.714

0.577–0.935 *

1.302

1.000–1.620

0.802

0.592 − 0.086

NX

NC

NC

NC

NC

NC

NC

  1. NC: not computable; CI: Credibility interval; TR: Time Ratio
  2. Deviance information criterion (DIC = 1633), Logarithm of the pseudo marginal likelihood (LPML= -811)
  3. The frailty component was significant in the multivariable model (Variance of frailties: 0.245, 95% CI (0.209–0.281))
  4. Trend effect: The model considered the trend effect for ordinal categorical variables. The variables BMI category, cancer site, surgery, radiotherapy, chemotherapy, morphology, disease stage could not be entered in the model in the multivariable model (All P > 0.05)
  5. *: P < 0.05