Table 4.
Full adjusted multivariate logistic regression model for predictors of EBF at 4–8 weeks post partum, 2010–2013
| Variable | Adjusted OR | 95% CI* |
| Year (ref: 2010) | ||
| 2011–12 | 2.08 | 1.87 to 2.32 |
| 2012–13 | 5.51 | 4.94 to 6.15 |
| Province (ref: EC) | ||
| FS | 1.22 | 0.99 to 1.50 |
| GP | 1.70 | 1.38 to 2.09 |
| KZN | 2.15 | 1.76 to 2.64 |
| LP | 0.97 | 0.77 to 1.22 |
| MP | 1.63 | 1.31 to 2.03 |
| NC | 1.33 | 1.07 to 1.66 |
| NW | 1.49 | 1.21 to 1.84 |
| WC | 1.30 | 1.03 to 1.64 |
| Mother age, years (ref: <20 years) | ||
| 20–34 | 1.07 | 0.99 to 1.61 |
| 35–50 | 1.12 | 1.01 to 1.25 |
| Mother education (ref: no education) | ||
| Grade 1–7 | 1.00 | 0.83 to 1.21 |
| Grade 8–12 | 0.86 | 0.72 to 1.03 |
| Tertiary | 0.62 | 0.50 to 0.77 |
| Parity (ref: primipara) | ||
| Multipara | 1.16 | 1.09 to 1.23 |
| Mother employment (ref: employed) | ||
| Other income | 1.53 | 1.42 to 1.65 |
| No income | 1.53 | 1.02 to 2.31 |
| HIV status (ref: negative) | ||
| Positive | 0.66 | 0.61 to 0.71 |
| Do not know | 0.74 | 0.56 to 0.99 |
| House type (ref: brick) | ||
| Informal/Wood | 1.09 | 1.00 to 1.19 |
| Traditional/Mud | 1.20 | 1.01 to 1.42 |
| Water source (ref: piped in house/yard) | ||
| Not piped in house/yard | 0.84 | 0.75 to 0.93 |
| Toilet (ref: indoor flush) | ||
| Pit latrine | 1.19 | 1.07 to 1.32 |
| Portable | 0.89 | 0.59 to 1.33 |
| Other | 1.55 | 0.80 to 3.02 |
| None | 1.03 | 0.82 to 1.29 |
| Fuel for cooking (ref: electricity) | ||
| Wood/Coal | 0.85 | 0.74 to 0.98 |
| Other | 0.96 | 0.57 to 1.62 |
| Food insecurity (ref: yes) | ||
| No | 1.13 | 1.04 to 1.23 |
| Do not know | 1.65 | 1.20 to 2.25 |
| Planned pregnancy (ref: yes) | ||
| No | 0.93 | 0.88 to 0.99 |
| Delivery mode (ref: vaginal) | ||
| Caesarean | 0.84 | 0.78 to 0.90 |
| Infant age, weeks (ref: 4 weeks) | ||
| 5 | 0.82 | 0.66 to 1.02 |
| 6 | 0.92 | 0.75 to 1.24 |
| 7 | 0.86 | 0.71 to 1.05 |
| 8 | 0.69 | 0.55 to 0.86 |
| Antenatal breastfeeding counselling (ref: yes) | ||
| No | 0.74 | 0.63 to 0.88 |
| Antenatal breastfeeding counselling by year (ref: yes in 2010)† | ||
| No ANC counselling in 2011–12 | 0.62 | 0.49 to 0.80 |
| No ANC counselling in 2012–2013 | 0.66 | 0.52 to 0.84 |
*Bold indicate 95% CI does not cross null (1.0); total observations: 29 288.
†This measures the multiplicative interaction of ANC breastfeeding counselling by year, hence the referent is a combination of having ANC breastfeeding counselling compared with the combination of no ANC breastfeeding counselling in each subsequent year.
ANC, antenatal care; EBF, exclusive breast feeding.