l ward or New Born Special Care Unit of the hospital.
2.1. Study Participation and Data Collection
Questionnaire for the study was adapted from previous similar studies   and was pretested prior to the commencement of the study among a sample of 20 mothers of which 95% found the content unambiguous.
Relevant information included in the study questionnaire were: maternal age, marital status, maternal and paternal education, maternal and paternal occupation, maternal birth characteristics (parity, mode of delivery―vaginal or caesarean section, antenatal care attendance in the index pregnancy, and maternal medical/psychological illness in the index pregnancy.
Also the social and medical characteristics of the newborns were collected including: infant’s post-natal medical condition, gestational age at birth, gender of baby, birth weight, and history of admission to the New-born Special Care Unit and the indication (s) for the hospitalization thereof.
The social class of each child was assessed using the criteria proposed by Oyedeji  in Ilesha, Nigeria with classification into five classes (social classes I to V). Here the social class of each child was determined based on the occupational status and highest educational attainment of the parents. The scores were summed and the mean (approximated to the nearest whole number) obtained. The mean score was used to assign the child to one of the socio-economic classes (1 to V). The social classes of the subjects were further stratified into upper (social classes I and II), middle (social class III) and lower (social classes IV and V).
2.2. Definition of Term
Early Initiation of Breastfeeding
Correct breastfeeding initiation time was regarded as maternal initiation of breastfeeding within 1 hour post-delivery   .
2.3. Statistical Analysis
Data were analyzed using the statistical package for Social Sciences, (SPSS) Windows Programme (SPSS Inc, Chicago, IL, USA) version 20.0. Chi square test was used to compare the proportions. Bivariate analysis was done and variables that showed significant level of less than 20 (i.e. p < 0.20) were in-put into the regression model.
The multivariate logistic regression model was used to determine the factors among parity, mode of delivery, post-natal medical history of newborns, planned pregnancy, maternal medical illness in the index pregnancy, gestational age at birth, birth weight, infant feeding decision made prior to conception and first food given to the newborn that actually predicted correct breastfeeding initiation within one hour of delivery, with the level of statistical significance set at p < 0.05.
A total of three hundred and twenty-one (321) mother and neonatal pairs were recruited and interviewed.
One hundred and fifty nine (49.5%) of the mothers were within the 20 to 29 years age bracket, and 315 (98.1%) were married.
Two hundred and eighteen respondents (67.9%) had tertiary education while 84 (26.2%) had at least secondary education. Most of the mothers belonged to social classes IV and V, accounting for 135 (42.1%) of the total respondents as shown in Table 1.
In Table 2, 181 (56.4%) of the respondents were multiparous mothers while 128 (39.9%) were primiparous. 312 (97.2%) respondents had adequate attendance at antenatal clinic/care in the index pregnancy while 190 (59.2%) were delivered vaginally.
Table 1. Socio-demographic characteristics of study subjects.
Table 2. Maternal birth characteristic.
Most of the mothers were stable in the post-delivery period with no significant medical illnesses reported in 252 (78.5%).
Two hundred and seventy (84.1%) of the newborns studied were medically stable while 51 (15.9%) were admitted into the NBSCU for different medical conditions post-partum.
One hundred and seventy five (54.5%) were males and majority of the neonates were of normal birth weight, 281 (87.5%) weighed up to 2.5 kg or more.
In general 86 (26.8%) of the respondent mothers correctly indicated breastfeeding within 1 hour of delivery.
Various indications for admission of the 51 sick neonates to the NBSCU were: perinatal asphyxia 21 (41.2%) neonatal jaundice 12 (23.5%), prematurity/low birth weight, 9 (17.6%), neonatal infections 5 (9.8%) and respiratory distress syndrome 4 (7.8%) as shown in Table 3.
On the test of association between early breastfeeding initiation time and the various maternal and newborn characteristics: parity (p = 0.000), mode of delivery (0.000), postnatal medical history of newborn (p = 0.001), breastfeeding as first food given to the newborn and the decision of the infant feeding made by the couple/mother pre-pregnancy showed statistical significance (p < 0.05) (Table 4).
Further statistics using bivariate analysis was done. The maternal and newborn characteristics/variables whose bivariate analysis showed a significant level
Table 3. Clinical characteristics of the newborns.
Key: NBSCU: Newborn Special Care Unit; LBW: low birth weight.
Table 4. Association between breastfeeding initiation time and maternal and newborn characteristics.
*EBF: Exclusive breastfeeding as defined by the World health organization is feeding of the infant only breast milk and no other liquids or solids except for drops or syrups consisting of vitamins, mineral supplements, or medicines  .
of less than 20 percent (i.e. p < 0.20) were in-put into the multivariate logistic regression model.
Primiparity, vaginal delivery, planned pregnancy, breastfeeding as first feeding given and infantfeeding decision made before pregnancy added significantly to the model/prediction table (Table 5).
The odds of having the correct time of initiation of breastfeeding is 5.3 times greater for breastfed babies as opposed to formula fed/other fed babies. The odds of commencing breastfeeding on time was also 2.67 times greater for those
Table 5. Predictors of correct breastfeeding initiation time among study babies.
babies where infant feeding decision was made before pregnancy and 1.9 times higher for primiparous women when compared with multiparous women. The odds of having correct initiation time of breastfeeding was 6.13 times less for babies delivered by Caesarean section or instrumental delivery and 2.2 times less in situations where the pregnancy was not planned.
In other words, breastfeeding of babies, making decision on infant feeding options before pregnancy and primparity were associated with an increase in having the correct time of initiation of breastfeeding. Conversely, assisted delivery (Caesarean section or instrumental delivery) and unplanned pregnancy were associated with a reduction in the likelihood of having a correct commencement time of breastfeeding.
From the current study, 86 (26.8%) out of the 321 mothers studied correctly initiated breastfeeding within one hour of delivery as recommended by the UNICEF/WHO. This rate is low as it falls within the zero to 29 percent bracket adjudged by the WHO as low  . It is also well below the Nigerian national prevalence of 40 percent in babies born in urban areas as well as 29 percent among those delivered in rural areas  . This prevalence is also much lower than the reports from some countries in sub-Saharan Africa and other developed countries including Kenya, Gambia, Cambodia, New Zealand and Italy     .
This reported low rate of correct breastfeeding initiation time as reported in the current study is coming at a time when the Nigerian recorded rate of 17% is regarded as one of the least exclusive breastfeeding globally  .
It has been reported that the low rates of correct BFI time and EBF in Nigeria could be addressed by complementing and sustaining the BFHI with Baby Friendly Community Initiative (BFCI) as has been demonstrated in Kenya which currently has one of the highest EBF rate of 97%  . However, it is also important to note that a lot of inconsistent definitions of correct breastfeeding initiation time abound in the literature making comparison of the standard timing unreliable  .
Timely BFI in newborns has been shown to have protective effects on the neonatal morality index within the first 28 days of life including overall neonatal mortality rate; infections related deaths, and low birth weights among other benefits as well as maternal health  .
The WHO has recommended that all mothers should initiate timely breastfeeding, with health workers assisting mothers to ensure that newborn babies are placed in skin-to-skin contacts with their mothers within one hour of birth in order to achieve this goal. Also mothers should recognize the cues of their newborns readiness to breastfeed. All of these efforts are geared towards successful and timely initiation and sustenance of exclusive breastfeeding    .
From the current study various maternal and newborn birth characteristics were significantly found to impact on timely breastfeeding initiation―planned pregnancy, parity, infant decision made before pregnancy, mode of maternal delivery, neonatal medical illness and breastfeeding as first food given to the newborn.
Currently, it appears that the efforts of the BFHI which targets the mothers presenting to the designated health facilities is not meeting up with the desired goals in our setting. Hence, complementing the BFHI with Baby Friendly Community Initiative (BFCI) may be the step in the right direction for many developing nations including Nigeria struggling to achieve this goal.
Such collaboration has resulted in huge EBF success elsewhere as the BFCI is community driven with local involvement, training of mothers’ support groups in the community with close links to the health facilities and applying messages obtained from the community interviews of the residents  .
Mothers who planned their pregnancies as well as made the decisions to breastfeed their babies prior to achieving pregnancy were more likely to correctly initiate BF on time if given the right family and health facility support as reported in the current study. Various workers working independently had reported that mothers who planned and made decisions to breastfeed their babies are more likely to initiate breastfeeding early     .
Some researchers had shown that the support a mother receives from health workers can influence her decision on whether or not to initiate breastfeeding timely. Forster and colleague  , showed that mothers should be encouraged by health workers to breastfeed immediately after birth, while efforts should be made to rightly identify those with higher risk of not successfully initiating and continuing breastfeeding and offer them extra support.
From the current study the odds of correctly initiating breastfeeding timely among mothers was 1.9 times higher for primiparous compared with multiparous mothers. The relationship between parity and birth order and breastfeeding has been found to be inconsistent among previous studies    . While some workers   reported that primiparous mothers were more likely to initiate EBF timely and more likely to continue breastfeeding; other researchers documented the reverse, pointing that multiparity was significantly associated with early/correct initiation and sustained breastfeeding possibly from their past experiences   , while Ekstrom et al.  and Guo and co-workers  found no associations.
Also the odds of having correct initiation time of breastfeeding was 6.13 times less for babies delivered by caesarean section or instrumental delivery compared to those delivered vaginally. Similar findings have been corroborated by previous workers   . Operative delivery has great psychological impact not only limited to the patient and her family, but sometimes on the attending physician who could become less motivated to encourage the mother to initiate breastfeeding early combined with the effects of anaesthesia on both the mother as well as on the newborn   . However, this is in contrast to mothers who had vaginal deliveries, most of whom had demonstrated early and timely initiation of breastfeeding of their newborns as the process of normal vaginal delivery increases the physiological stimulation of milk production and therefore are more likely to have enough milk and initiate breastfeeding early  .
Various maternal as well as newborn medical conditions could adversely affect early initiation of breastfeeding in the newborns. Such factors include prematurity, low birth weight, infections, congenital abnormalities, prenatal asphyxia among others. Mother with healthy babies are more likely to initiate breastfeeding timely compared with those who had sick newborns   .
Mothers who decided to practice exclusive breastfeeding before achieving pregnancy started their babies on breastmilk first and early as observed in the current study. Mothers who intended or practiced pre-lacteal feeding with formula milk, glucose water or whatever in their newborns first were more likely not to initiate breastfeeding early as it is when the newborn begins to suckle the maternal nipples that proper breast milk let down is established    .
Maternal education as well as social class was not significantly associated with timely initiation of breastfeeding in the current study. This is in contrast to the findings by previous workers    , that women with higher educational status were more likely to initiate and continue breastfeeding in their infants. This could partly be due to the fact that majority of the respondents in the current study had higher educational attainment.
In conclusion, the timely breastfeeding initiation rate of 26.8 percent among mothers in the current study is adjudged to be low in accordance with the World Health Organization standards.
Planned pregnancy, making decision of infant feeding option before pregnancy, primiparity, vaginal delivery, breastfeeding as the first food given to the neonate after delivery were associated with having the correct timing of breastfeeding initiation among respondents.
Efforts should be made to strengthen BFHI and complement it with Baby Friendly Community Initiative in order to improve the rate of correct breastfeeding initiation time in our setting to save lives.