Table 4 Univariate and multivariate covariate-specific HR for surgery vs. active monitoring for prostate cancer.

From: Uncovering interpretable potential confounders in electronic medical records

 

Univariate analysis

Multivariate analysis

Covariates

HR

95% CI

P value

HR

95% CI

P value

W.surgery

1.67

[0.99, 2.8]

0.057

1.02

[0.55, 1.9]

0.957

struct:patient_age*

3669.74

[5.3e+02, 2.5e+04]

<0.001

143.94

[11, 1.9e+03]

<0.001

struct:race_white

0.87

[0.41, 1.8]

0.709

0.68

[0.23, 2]

0.478

struct:race_api

0.59

[0.15, 2.3]

0.443

0.81

[0.16, 4.1]

0.799

struct:race_black

2.04

[0.41, 10]

0.384

5.16

[0.82, 32]

0.080

struct:hispanic

1.09

[0.29, 4.1]

0.898

1.68

[0.41, 6.8]

0.471

struct:clinical_stage

2.58

[0.31, 21]

0.378

3.75

[0.35, 40]

0.275

struct:tumor_grade

0.22

[0.014, 3.7]

0.296

2.37

[0.0084, 6.6e+02]

0.764

struct:grade_unknown

0.06

[0.00094, 4.2]

0.198

0.02

[2.5e-05, 14]

0.235

struct:diagnosis_year

0.12

[0.027, 0.55]

0.006

0.50

[0.073, 3.4]

0.483

text:bladder*

160.34

[65, 3.9e+02]

<0.001

24.17

[6.6, 89]

<0.001

text:urothelial*

2178.75

[5e+02, 9.6e+03]

<0.001

68.15

[7.4, 6.3e+02]

<0.001

  1. HR hazard ratio, CI confidence interval, * intersection terms.
  2. The * denotes intersection terms identified by our method. The lower block of covariates represents terms extracted from clinical notes. For each covariate, we show the effect size (HR), the 95% confidence interval (CI), and the statistical significance (P value) from a Wald statistics test.
Back to article page