Table 4.
Bivariate analysis and logistic regression model predicting sexual onset among YHH in Gikongoro at the time of the survey
| Bivariate n = 692 |
Multivariateφ n = 692 |
|||
|---|---|---|---|---|
| Odds Ratio | 95% CI | Odds Ratio | 95% CI | |
| Demographic characteristics | ||||
| Age | ||||
| 12-18 | 0.42*** | 0.28-0.62 | 0.60* | 0.37-0.98 |
| 19-24 | Ref. | Ref. | ||
| Gender | ||||
| Male | 1.48** | 1.07-2.04 | 1.30 | 0.89-1.91 |
| Female | Ref. | Ref. | ||
| Orphan Status | ||||
| Double | Ref. | Ref. | ||
| Maternal | 0.85 | 0.57-1.25 | 0.73 | 0.46-1.15 |
| Paternal | 0.98 | 0.46-2.07 | 1.04 | 0.44-2.44 |
| Both Alive | 0.94 | 0.17-5.20 | 0.86 | 0.12-5.94 |
| Uncertain | 0.82 | 0.33-2.00 | 0.67 | 0.24-1.86 |
| Number of years as head of household | ||||
| < 5 | 0.66** | 0.48-0.91 | 0.78 | 0.54-1.15 |
| ≥ 5 | Ref. | Ref. | ||
| Number of other children & adolescents in household | ||||
| 0 | 1.20 | 0.54-2.64 | 1.53 | 0.59-3.93 |
| 1-2 | 1.69 | 0.80-3.56 | 2.07 | 0.87-4.89 |
| 3-4 | 1.13 | 0.52-2.44 | 1.34 | 0.55-3.25 |
| 5 or more | Ref. | Ref. | ||
| School attendance | ||||
| Ever attended school (Yes versus No) | 1.63 | 0.97-2.74 | N/A | |
| Currently in school (Yes versus No) | 0.40** | 0.20-0.76 | 0.51 | 0.25-1.05 |
| Education completed | ||||
| None | 1.18 | 0.61-2.28 | N/A | |
| Primary (7 years) | 0.81 | 0.39-1.68 | N/A | |
| More than primary | Ref. | |||
| Household characteristics | ||||
| Eat one meal per day | 1.29 | 0.94-1.78 | N/A | |
| Assets index (continuous) | 0.91 | 0.82-1.01 | N/A | |
| Social connectedness | ||||
| Belong to 1 or more community groups (Yes versus No) | 0.84 | 0.61-1.17 | N/A | |
| Peers relationship | ||||
| One close friend they can count on | 0.55* | 0.32-0.94 | 0.51* | 0.28-0.94 |
| Feel they belong to a group of friends | 0.79 | 0.46-1.38 | N/A | |
| Moved 2 or more times during last 5 years (Yes versus No) | 1.89*** | 1.34-2.67 | 1.85** | 1.22-2.79 |
| Adult support (≤ 3 versus > 3) | 1.11 | 0.79-1.54 | N/A | |
| Youth behaviors and emotional distress | ||||
| Drink alcohol | 1.31 | 0.95-1.80 | N/A | |
| Use drugs | 14.66** | 1.79-119.9 | 9.77* | 1.14-83.44 |
| Depression (Cut-off = 24) | 1.39* | 1.01-1.91 | 1.14 | 0.78-1.65 |
| HIV/AIDS Prevention Knowledge | ||||
| Abstain from sex | 0.82 | 0.19-3.46 | N/A | |
| Condom use | 1.68** | 1.21-2.34 | 1.22 | 0.83-1.79 |
| Limit sex to one partner | 0.99 | 0.69-1.45 | N/A | |
| Avoid traditional doctors | 1.07 | 0.77-1.49 | N/A | |
| Perception of HIV infection risk | ||||
| No risk/possibility | Ref. | Ref. | ||
| Small/Moderate/Great risk | 2.19*** | 1.56-3.07 | 1.95*** | 1.32-2.86 |
*p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001
φFive basic socio-demographic variables (age group, gender, orphan status, number of years as HH, number of other children & adolescents in household) and variables that attained P < 0.01 in the bivariate analyses were entered into the logistic regression model. -N/A = Not Applicable; Ref. = Reference group; CI = Confidence Interval