Skip to main content

Table 3 Covariates associated with “heavy episodic drinking” in Kenya

From: Prevalence and determinants of heavy episodic drinking among adults in Kenya: analysis of the STEPwise survey, 2015

 

Unadjusted Odds Ratio (95% CI)

P-value

Adjusted Odds Ratioa (95% CI)

P-value

Age (per 10 years)

1.15 (1.03,1.29)

0.01

1.15 (0.98,1.34)

0.08

Age categories

 

0.10

  

 18–29

1.0

   

 30–39

1.7 (1.1,2.7)

0.02

  

 40–49

1.9 (1.0,3.5)

0.05

  

 50–59

1.2 (0.8,1.8)

0.46

  

 60–69

1.7 (1.0,3.0)

0.07

  

Sex

    

 Men

9.9 (5.3,18.8)

<.0001

  

 Women

1.0

   

Sexacurrently smoking

   

0.006

 Smoker subgroup: man vs. woman

  

2.0 (0.7,5.3)

0.19

 Non-smoker: man vs. woman

  

7.9 (4.1,15.5)

< 0.0001

Marital status

 

0.31

 

0.26

 Currently married/ Cohabiting

1.0

 

1.0

 

 Never married

1.2 (0.8,1.8)

0.44

0.9 (0.6,1.4)

0.66

 Formerly married/widowed

1.4 (0.8,2.5)

0.19

1.8 (0.9,3.5)

0.10

Education level

 

0.12

 

0.50

 No education

1.0

1.0

 Primary

1.6 (0.9,2.9)

0.11

1.2 (0.6,2.3)

0.57

 Secondary

2.0 (1.04,3.9)

0.04

1.5 (0.8,2.8)

0.21

 Tertiary

2.5 (1.1,5.6)

0.02

1.6 (0.7,3.8)

0.28

Wealth quintile

 

0.02

 

0.02

 Poorest

1.00

 

1.0

 

 Second

1.0 (0.5,1.9)

0.92

0.8 (0.4,1.6)

0.45

 Middle

1.0 (0.5,2.0)

0.90

0.7 (0.4,1.5)

0.38

 Fourth

1.2 (0.6,2.4)

0.62

0.8 (0.4,1.8)

0.64

 Richest

1.9 (0.9,4.1)

0.07

1.7 (0.8,3.8)

0.18

Residence

    

 Rural

0.6 (0.4,1.0)

0.04

1.0 (0.7,1.5)

0.86

 Urban

1.00

   

Currently smoking

    

 Yes

6.9 (4.4, 10.8)

<.0001

  

 No

1.00

   
  1. aThe final model (adjusted model) included age, marital status, education, wealth quintile, residence, gender, currently smoking, and interaction of gender and currently smoking. The interaction of each predictor with smoking status for HED outcome was tested, however only interaction of gender by smoking remained significant. Because interaction term of gender by smoking is significant, the main effects of smoking and gender are not presented. Instead, gender effect is presented stratified by smoking status (the interaction indicates that the effect of gender differs significantly by smoking status)