Skip to main content

Table 3 Bangladesh logistic regression models (Best AIC) by cause

From: On the estimation of population cause-specific mortality fractions from in-hospital deaths

Cause

Parameter

Age†

Sex‡

Disabled

Male HH

W/S/D

Municipality

Intercept

Cancer

OR

–

–

0.85

–

–

–

–

Beta

–

–

− 0.64

–

–

–

− 2.39

Diabetes

OR

0.79

–

0.81

–

–

1.25

–

Beta

− 0.37

 

− 0.77

–

–

0.92

0.3

Acute infections

OR

0.83

–

0.67*

–

0.66*

–

–

Beta

− 0.21

–

− 1.5

–

−1.54

–

− 0.12

Chronic infections

OR

0.80

–

–

–

–

–

–

Beta

− 0.31

–

–

–

–

–

− 0.32

Injuries

OR

–

–

0.47*

0.84

0.79

1.46*

–

Beta

–

–

− 3.12

− 0.63

− 1.01

1.71

− 1.15

Maternal

OR

0.66

–

–

–

–

–

–

Beta

− 0.89

–

–

–

–

–

1.94

Other NCDs

OR

0.79*

0.91

0.88*

–

0.84*

–

–

Beta

− 0.3

− 0.37

− 0.49

–

− 0.68

–

0.75

Stroke

OR

0.87*

–

0.70*

–

–

–

–

Beta

− 0.21

–

− 1.31

–

–

–

− 0.21

  1. Abbreviations: NCDs non-communicable diseases, HH household, W/S/D widowed/separated/divorced, OR odds ratio, CI confidence interval
  2. *(significant at the 0.05 level), † (decimal decades), ‡ (1 = Female, 2 = Male)