Skip to main content

Table 2 Heckman Selection Model (N = 1605)

From: Assessing response bias from missing quality of life data: The Heckman method

  

95% Confidence Interval

Variables

Coefficient

Lower Limit

Upper Limit

Mean patient age

0.01

0.00

0.02

Male gender

0.44

-0.07

0.95

Caucasian race

-0.06

-0.25

0.13

Hx of MI

0.02

-0.16

0.20

Hx of Chronic heart failure

-0.01

-0.20

0.18

Hx of Coronary artery bypass graft surgery

0.09

-0.13

0.31

Hx of Percutaneous coronary intervention

0.28

0.01

0.54

History of trauma**

-0.29

-1.37

0.80

Substance or alcohol abuse

-0.21

-0.40

-0.01

Current smoker

-0.10

-0.29

0.09

Dementia

0.02

-0.76

0.79

COPD

0.23

0.05

0.41

Stroke

-0.12

-0.39

0.15

Depression

-0.26

-0.48

-0.05

Arthritis

0.08

-0.07

0.23

Diabetes mellitus

-0.02

-0.25

0.21

Peptic ulcer disease**

0.04

-0.20

0.28

Coronary revascularization during index admission

0.12

-0.06

0.30

Cardiogenic shock

0.04

-0.81

0.89

Hypotensive episode

0.05

-0.29

0.38

Do not resuscitate order**

-0.68

-1.09

-0.27

Positive stress test

-0.01

-0.25

0.24

Admitted to Tertiary VA

-0.09

-0.28

0.10

ST segment elevation MI on ECG

-0.04

-0.29

0.22

Mean serum creatinine

0.01

-0.06

0.07

Left bundle branch block on ECG

-0.02

-0.41

0.38

Cardiac catheterization during index admission

0.04

-0.10

0.19

Discharge diagnosis Unstable Angina (vs. MI)

0.06

-0.20

0.32

  1. MI: myocardial infarction; Hx: History; COPD: chronic obstructive pulmonary disease; ECG: electrocardiogram;
  2. **Instruments not included in outcomes equations