Table 3 Multilevel logistic regression analysis assessing the effect of background characteristics on the likelihood of substance use among adolescent boys
Background characteristics | Model-1 | Model-2 | Model-3 |
|---|---|---|---|
OR (95% CI) | OR (95% CI) | OR (95% CI) | |
Age (in years) | |||
 Early adolescents (10–14) | Ref. | Ref. | Ref. |
 Late adolescents (15–19) | 7.38*(1.67,32.6) | 6.81*(3.07,15.07) | 6.65*(3.06,14.45) |
Current schooling | |||
 Never attended | Ref. | Ref. | Ref. |
 Dropout | 0.68 (0.35,1.34) | 0.74 (0.42,1.30) | 0.67 (0.37,1.20) |
 Currently attending | 0.08*(0.01,0.59) | 0.11*(0.04,0.32) | 0.10*(0.03,0.29) |
Working status | |||
 No | Ref. | Ref. | Ref. |
 Yes | 5.09*(1.63,15.94) | 4.22*(2.32,7.67) | 4.46*(2.42,8.22) |
Media exposure | |||
 No exposure | Ref. | Ref. | Ref. |
 Rarely | 0.88 (0.44,1.74) | 0.92 (0.49,1.73) | 0.89 (0.47,1.68) |
 frequently | 1.29 (0.67,2.5) | 1.40 (0.76,2.57) | 1.43 (0.78,2.63) |
Substance use among family members | |||
 No-one |  | Ref. | Ref. |
 Someone in the family |  | 2.22*(1.48,3.33) | 2.13*(1.44,3.17) |
Caste | |||
 Scheduled Caste/Scheduled Tribe |  | Ref. | Ref. |
 Non- Scheduled Caste/Scheduled Tribe |  | 0.75 (0.55,1.03) | 0.72*(0.52,0.99) |
Religion | |||
 Hindu |  | Ref. | Ref. |
 Non-Hindu |  | 1.09 (0.77,1.55) | 1.28 (0.88,1.84) |
Wealth index | |||
 Poorest |  | Ref. | Ref. |
 Poorer |  | 0.69 (0.42,1.15) | 0.72 (0.44,1.19) |
 Middle |  | 0.80 (0.50,1.30) | 0.90 (0.56,1.45) |
 Richer |  | 0.66 (0.40,1.08) | 0.78 (0.47,1.28) |
 Richest |  | 0.80 (0.49,1.33) | 1.05 (0.62,1.78) |
Community education | |||
 High |  |  | Ref. |
 Low |  |  | 0.78 (0.56,1.09) |
Community wealth | |||
 High |  |  | Ref. |
 Low |  |  | 1.31 (0.90,1.92) |
Community substance use | |||
 Low |  |  | Ref. |
 High |  |  | 1.24*(1.01,1.68) |
Residence | |||
 Urban |  |  | Ref. |
 Rural |  |  | 1.09 (0.76,1.54) |
State | |||
 Uttar Pradesh |  |  | Ref. |
 Bihar |  |  | 1.56*(1.12,2.17) |