Table 3 Univariate and multivariate covariate-specific HR for surgery vs. radiation 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

1.27

[0.77, 2.1]

0.352

1.09

[0.59, 2]

0.777

struct:patient_age*

594.88

[87, 4.1e+03]

<0.001

35.96

[3.5, 3.7e+02]

0.003

struct:race_white

0.92

[0.44, 1.9]

0.822

0.65

[0.22, 1.9]

0.439

struct:race_api

0.63

[0.18, 2.2]

0.467

0.67

[0.14, 3.3]

0.622

struct:race_black

1.63

[0.33, 8.1]

0.551

4.04

[0.64, 25]

0.137

struct:hispanic

0.85

[0.2, 3.6]

0.831

1.52

[0.33, 7]

0.593

struct:clinical_stage

0.30

[0.042, 2.2]

0.237

1.02

[0.14, 7.4]

0.987

struct:tumor_grade

0.05

[0.0013, 2]

0.111

0.10

[0.00038, 24]

0.406

struct:grade_unknown

0.55

[0.028, 11]

0.698

0.99

[0.0029, 3.4e+02]

0.996

struct:diagnosis_year

0.12

[0.024, 0.57]

0.008

0.17

[0.025, 1.2]

0.075

text:bladder*

207.51

[79, 5.4e+02]

<0.001

35.95

[9.3, 1.4e+02]

<0.001

text:urothelial*

1919.54

[4.2e+02, 8.7e+03]

<0.001

44.07

[4.4, 4.4e+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