Skip to main content

Table 3 Multilevel mixed effects logistic regression showing predictors of multiple sexual partnership among adolescent boys and young men between 2003 and 2014

From: Multiple sexual partnership among adolescent boys and young men in Ghana: analysis of the 2003–2014 Ghana Demographic and Health Survey

 

Model 0

(null model)

Model I

Model II

Model III

Model IV

ORadj (95% CI)

ORadj (95% CI)

ORadj (95% CI)

ORadj (95% CI)

Fixed effects results

Individual-level factors

Age

     

 15–19 (ref)

     

 20–24

 

1.43(1.05–1.94)*

  

1.39(1.01–1.91)*

Age at first sex

     

 < 15 years (ref)

     

 15–17 years

 

0.82(0.54–1.24)

  

0.83(0.54–1.26)

 18 + years

 

0.47(0.30–0.72)**

  

0.45(0.29–0.70)***

Educational attainment

 No education (ref)

     

 Primary

  

1.52(0.78–3.26)

 

1.77(0.84–3.73)

 Secondary or higher

  

2.23(1.14–4.33)

 

2.94(1.44–6.03)*

Knowledge of ovulation cycle

 

 Correct (ref)

     

 Wrong

 

1.18(0.86–1.60)

  

1.10(0.80–1.51)

Current contraceptive use

 

 No method (ref)

     

 Folkloric/traditionalist

 

0.61(0.35–1.06)

  

0.62(0.35–1.08)

 Modern method

 

0.39(0.29–0.52)***

  

0.37(0.28–0.50)***

Smokes cigarettes

 No (ref)

     

 Yes

 

1.66(0.71–3.89)

  

1.93(0.80–4.65)

Household-level factor

Wealth quintile

     

 Poorest (ref)

     

 Poorer

 

0.88(0.56–1.38)

  

0.78(0.49–1.25)

 Middle

 

1.02(0.67–1.56)

  

0.86(0.54–1.38)

 Richer

 

1.70(1.13–2.53)*

  

1.35(0.82–2.23)

 Richest

 

2.11(1.37–3.25)**

  

1.76(1.01–3.06)*

Interpersonal-level factors

 

Heard FP on radio

 No (ref)

     

 Yes

  

0.86(0.65–1.18)

 

0.89(0.65–1.21)

Heard FP on TV

 No (ref)

     

 Yes

  

1.09(0.82–1.47)

 

0.96(0.70–1.33)

Community-level factors

Religion

     

 Christians (ref)

     

 Muslims

   

0.87(0.60–1.27)

0.88(0.59–1.30)

 Traditionalist/spiritualist

   

1.54(1.05–2.27)

1.60(1.07–2.41)

 Other

   

1.17(0.73–1.88)

1.25(0.76–2.06)

Ethnicity

     

 Akan (ref)

     

 Non-Akans

   

0.85(0.65–1.12)

1.04(0.78–1.40)

Place of residence

     

 Urban (ref)

     

 Rural

   

0.72(0.55–0.93)*

0.98(0.69–1.38)

Structural-level

     

Year of survey

     

 2003 (ref)

     

 2008

 

1.87(1.31–2.66)**

1.98(1.39–2.80)***

1.94(1.37–2.76)***

1.691(1.17–2.44)***

 2014

 

2.28(1.59–3.28)***

2.19(1.54–3.13)***

2.19(1.54–3.12)***

1.97(1.35–2.87)***

Random effects result

 AIC

1696.482

1564.51

1604.276

1604.213

1551.36

 ICC

0.049

0.067

0.077

0.074

0.071

 LR test

4.83**

3.21*

4.68*

4.21*

3.15*

 Log likelihood

− 845.241

− 758.123

− 794.138

− 793.106

− 751.680

 Wald Chi-square

Reference

88.40***

30.36***

32.71***

97.90***

Observations

1422

1422

1422

1422

1422

  1. p value < 0.05 = *, p value < 0.01**, p value < 0.001***