Skip to main content

Table 6 MCS Model Corrected For Response Bias (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 (per 1 year increment)

0.06

0.00

0.12

Male gender

0.25

-2.65

3.15

Caucasian race

-0.51

-1.24

0.22

Hx of MI

-1.24

-2.07

-0.40

Hx of Chronic heart failure

0.17

-0.94

1.27

Hx of Coronary artery bypass graft surgery

-0.01

-1.21

1.18

Hx of Percutaneous coronary intervention

-0.19

-1.36

0.99

Substance or alcohol abuse

-1.55

-3.39

0.29

Current smoker

-0.17

-1.46

1.13

Dementia

-2.81

-6.90

1.27

COPD

-1.76

-2.80

-0.72

Stroke

-1.75

-4.25

0.75

Depression

-1.33

-2.71

0.04

Arthritis

-10.80

-12.83

-8.78

Diabetes mellitus

-0.35

-1.43

0.72

Coronary revascularization during index admission

0.70

-0.48

1.89

Cardiogenic shock

-2.58

-7.28

2.12

Hypotensive episode

-0.16

-2.26

1.94

Positive stress test

0.01

-2.24

2.25

Admitted to Tertiary VA

0.24

-1.10

1.58

ST-segment elevation MI on ECG

1.92

0.91

2.92

Mean serum creatinine (per 1 mg/dL increment)

-0.37

-0.97

0.24

Left bundle branch block on ECG

-2.00

-4.63

0.63

Cardiac catheterization during index admission

-0.46

-1.63

0.72

Discharge diagnosis Unstable Angina (vs. MI)

0.04

-1.23

1.32

Inverse Mills Ratio *

-8.93

-10.73

-7.13

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