Table 2.
Incidence and mortality of neonatal sepsis.
| Intervention (n = 1575) | Control (n = 1561) | RR (95%CI) | AdjRRa (95%CI) | |
|---|---|---|---|---|
| Suspected sepsis no. (%) | 361 (22.9) | 434 (27.8) | 0.82 (0.73–0.93) | 0.82 (0.73–0.93) |
| Sepsis related mortality no. (%) | 70/1575 (4.4%) | 109/1561 (6.9%) | 0.64 (0.48–0.85) | 0.63 (0.47–0.85) |
Suspected sepsis was defined as one or more of the following signs or symptoms: temperature of 35.5 °C or more than 38 °C, no movement or movement only on stimulation, in-drawing of the chest, and convulsions. Signs and symptoms were not reported for the first 24 h of life. After that time, the infant should have been well for at least 24 h before becoming sick. The denominator excludes infants who died before reaching 48 h of age.
Risk ratios (RR) were adjusted for clustering due to multiple births and in accordance with study site, type of delivery, multiple pregnancies, mother's age at randomization, infant's sex and weight, mother's years of schooling and age, household with toilet, and family income.