Table 3 Variable selection through the univariate/multivariate Cox-PH regression analysis for predicting risks of colorectal adenoma occurrence based on the primary cohort.

From: Derivation and validation of a nomogram incorporating modifiable lifestyle factors to predict development of colorectal adenomas after negative index colonoscopy

Variables

The univariate Cox-PH regression analysis

Variables

The multivariate Cox-PH regression analysis

HR (95% CI)

p-Value

HRa (95% CI)

p-Value

VIF

Age

1.099 (1.048–1.153)

0.001

Age

1.102 (1.047–1.161)

< 0.0001

1.040

BMI

1.147 (1.038–1.269)

0.007

BMI

1.133 (1.031–1.246)

0.010

1.003

Gender

 

0.025

Gender

 

0.272

 

Female

Reference

 

Female

Reference

 

Reference

Male

1.743 (1.073–2.832)

 

Male

1.325 (0.802–2.189)

 

1.043

Smoking status

 

0.197

Smoking status

 

 

Never

Reference

 

Never

 

Current or past

1.481 (0.816–2.691)

 

Current or past

 

Alcohol consumption

 

0.077

Alcohol consumption

 

 

Never

Reference

 

Never

 

Occasional, frequent or even daily

1.683 (0.945–3.000)

 

Occasional, frequent or even daily

 

Physical activity

 

0.438

Physical activity

 

 

Physical inactivity

Reference

 

Physical inactivity

 

Regular activity

0.822 (0.500–1.349)

 

Regular activity

 

Family history of CRC in FDR

 

0.255

Family history of CRC in FDR

 

 

No

Reference

 

No

 

Yes

1.590 (0.716–3.529)

 

Yes

 

History of chronic constipation

 

0.495

History of chronic constipation

 

 

No

Reference

 

No

 

Yes

1.203 (0.707–2.047)

 

Yes

 

History of chronic diarrhea

 

0.586

History of chronic diarrhea

 

 

No

Reference

 

No

 

Yes

0.861 (0.502–1.477)

 

Yes

 

History of chronic appendicitis or appendectomy

 

0.704

History of chronic appendicitis or appendectomy

 

 

No

Reference

 

No

 

Yes

1.179 (0.504–2.759)

 

Yes

 

  1. aHRs were adjusted for baseline age, BMI and gender.