Characteristics of respondents
A total of 98 mother and new-born pairs participated in the study. We had distributed 120 questionnaires and 98 were returned giving an overall response rate of 81.7%. There were 40 (40.8 %) new-borns clinically confirmed to have neonatal sepsis. The mothers interviewed were aged between 15 and 44 years with 44.9% in the age category 21 to 30. The majority (67.4 %; 66/98) of the mothers delivered through normal vaginal and 66.2% were married. Majority of the babies 62.6% (57/98) had a birth weight above 2500g. There were more female new-borns (52.3 %) than males. Table 1 shows the maternal and neonate characteristics.
Antenatal clinic visits and gestational age at first visit
Figure 1 shows the antenatal clinic (ANC) service providers for the participants. Out of those who attended ANC in Public clinic/hospital, 42.7% (38/87) indicated that they had attended 3-4 times and 27.6% (24/87) indicated that they had attended more than 4 times. The majority of the participants 64.3% (63/98) were at 4 to 6 months gestational age at first ANC visit (Figure 2).
New-born care practices
New-born care practices reported included hand washing practices (Figure 3), cord cleaning frequency (Figure 4) and material used for cleaning. Ninety-four percent (80/98) of the respondents indicated that they use methylated spirit and cotton wool to clean the baby’s cord while 5.9% (5/98) indicated that they had used breast milk for cleaning (Table 2). Figure 5 shows the sources of information on new-born care for the respondents. The majority of the respondent’s 62.2(/98) indicted health care workers as their main source of information on new-born care, followed by the mother/relative (43.3%), and social media constituting 11.1%.
Neonate factors associated with neonatal sepsis
Sixty per cent of sepsis cases were among new-borns with low birth weight. Results show that 30% of the new-born babies born with normal birth weight were protected from sepsis OR=0.7 (95% CI: 0.3-1.7), p=0.45 as compared to those born with low birth weight. However, there was no statistical significance as shown in Table 3. Sepsis was not in any way associated with the person who conducted delivery, with cases of sepsis prevalent in-home delivery, nurse/midwife and doctor. A significant association was found (p=0.02), with 61% reduction in sepsis cases among new-born babies who did not experience problems before, during or after delivery OR=0.39(95% CI: 0.16-0.90), p=0.03. Stepwise multivariable logistic regression showed that among new-borns without problems before, during and after delivery 56 % of the babies were less likely to have sepsis OR=0.44 (95% CI: 0.18-1.1), p=0.07. The common problems experienced are summarized in Table 4.
Maternal factors associated with neonatal sepsis
Mothers age (p=0.02) and the number of ANC visits (p=0.04) were associated with neonatal sepsis (Table 5). Sixty-six per cent of the new-born babies with protection from sepsis were from mothers who have attended ANC 1 to 2 times compared to those who did not attend adjusting for other factors, OR=0.34 (95% CI:0.1-1.2). Ninety per cent of the new-born from mothers who had visited ANC 3 to 4 times were less likely to develop sepsis compared to those from mothers who had never attended.
New-born care practices and risk of neonatal sepsis
A high proportion of sepsis cases were seen among new-borns whose mothers practiced inappropriate cord care such as no strict hand washing before handling the baby, cord care and before breastfeeding. However, the differences among these groups weren't statistically significant (Table 6). Breast milk application was associated with sepsis in the univariate analysis (p=0.01), 79% of the mothers who did not apply breast milk to the cord were less likely to develop sepsis OR=0.22(95%CI: 0.1-0.78). odds of Mothers who got information from other sources not health care workers had 2.61 times increased developing sepsis compared to those who received information from health care workers. Not washing hands before cord care, applying cow dung, applying breast milk, applying other substances, getting information about cord care from health care and mother/relative were variables that were identified to be associated with sepsis development (univariate analysis with a cut-off point of p-value<0.20). However, these variables were selected as candidates of the multivariable model. Getting information from health care workers was associated with decreased odds of sepsis development OR=0.22(95% CI: 0.10-0.87) and applying breast milk on cord had an increased risk of sepsis OR= 2.4(95% CI: 0.9-6.2) but a midwife at the health facility noted that most mothers were influenced by local traditional practices to apply different remedies to the cord of the new-born thus predisposing them to infection.
Mothers’ knowledge of signs associated with neonatal illness
Figure 6 shows reported signs of neonatal illness by the respondents. Majority of the respondents (64.4%) reported fever, 44.8% reported refusal to breastfeed and only 20% reported convulsions.
Interviewed health care worker perceptions
Demographic characteristics of 36 health care workers that were interviewed are indicated in Table 7. Shortage of medical sundries such as methylated spirit, hand sanitizers and surgical gloves was indicated as a major challenge by the majority of respondents (81%). Inadequate supply or range of antibiotics was also reported. P29 reported that:
“At times there is no soap to wash hands, babies share incubators and rooms are crowded.”
The lack of training or refresher programmes on infection control was reported by of the participants. One of the interviewed cases (P2) said:
“Such workshops are there but the same individuals are the ones who attend.”
There was a strong feeling among participants that routine screening for infections such as HIV and syphilis during pregnancy was being done at the institution ANC but sometimes there is a lack of resources which may lead to some infections being missed.
Cord care was cited by the interviewed participants as the most common information on prevention of neonatal sepsis given to mothers after delivery. Mothers were taught how to clean the cord with methylated spirit every 3 to 4 hours daily until the cord stump falls off. Mothers were also taught the importance of exclusive breastfeeding for at least six months and to report early when a baby is not feeling well. Mothers were also encouraged to wash hands regularly before breastfeeding, before cord care, before touching the baby and when changing nappies. Mothers were also taught to clean feeding bottles before using them and encouraged to wash and iron babies’ clothes regularly.