Skip to main content

Table 3 Initial PCS Model

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 (per 1 year increment)

-0.06

-0.11

-0.01

Male gender

-1.04

-4.31

2.23

Caucasian race

-0.69

-1.72

0.35

Hx of MI

-0.85

-2.46

0.75

Hx of Chronic heart failure

-2.96

-4.36

-1.55

Hx of Coronary artery bypass graft surgery

-3.12

-4.29

-1.95

Hx of Percutaneous coronary intervention

-1.09

-2.68

0.51

Substance or alcohol abuse

-0.30

-1.89

1.30

Current smoker

-0.57

-1.66

0.52

Dementia

-0.84

-3.17

1.50

COPD

-3.76

-4.77

-2.76

Stroke

-3.01

-4.34

-1.67

Depression

-4.61

-6.07

-3.15

Arthritis

-1.89

-2.94

-0.83

Diabetes mellitus

-2.33

-3.56

-1.09

Coronary revascularization during index admission

1.91

0.38

3.44

Cardiogenic shock

1.62

-3.88

7.12

Hypotensive episode

0.13

-1.93

2.19

Positive stress test

1.15

-0.47

2.76

Admitted to Tertiary VA

-0.88

-2.00

0.24

ST-segment elevation MI on ECG

2.34

0.81

3.86

Mean serum creatinine (per 1 mg/dl increment)

-0.64

-1.13

-0.16

Left bundle branch block on ECG

0.09

-2.43

2.61

Cardiac catheterization during index admission

0.63

-0.45

1.71

Discharge diagnosis Unstable Angina (vs. MI)

-0.66

-1.54

0.21

Inverse Mills Ratio *

   
  1. * Inverse Mills Ratio: Variable derived from the Heckman Selection Equation; MI: myocardial infarction; Hx: History; COPD: chronic obstructive pulmonary disease; ECG: electrocardiogram