Neonatal mortality and perinatal risk factors in rural south western Nigeria: a community-based prospective study

ABSTRACT

BACKGROUND: Reliable data on births and deaths particularly at the community level are scarce yet they are urgently needed to inform policy and assess the improvements which may haveoccurred with recent interventions. OBJECTIVE: To determine neonatal mortality rate and identify perinatal risk factors associated with neonatal deaths. METHODS: In a community-based prospective study, baseline data on births and deaths were collected as they occurred in a rural community of Southwest Nigeria from 1993 to 1998. Data on births and deaths were collected for the period. RESULTS: There were 972 live births and 64 infant deaths giving an infant mortality rate of 65.8 per 1000. Neonatal deaths accounted for a half of all infant deaths (N=32) giving a neonatal mortality rate of 32.9 per 1000. Twelve (37.5%) of neonatal deaths occurred on the first day of life; half of all neonatal deaths occurred within two days of birth, 21(65.6%) occurred during the first seven days of life and only 11 (34.4%) occurred over the last three weeks of the first month. The commonest known cause of death was associated with low birth weight (LBW) which was responsible for eight (25%) of deaths, while sepsis/fever and maternal deaths/failure to thrive were responsible for four (12.5%) and three (9.4%) deaths respectively. Asphyxia accounted for 3(9.4%) deaths; neonatal tetanus, congenital abnormality and diarrhoea were responsible for one (3.1%) death each. Cause of death was unclassified in many early neonatal deaths particularly those which occurred at home. Predictors of neonatal death included LBW {RR=4.7 (1.7-13.1) p=0.03},delivery outside a health facility {RR=3.6 (1.001-13.2) p=0.05},lack of attendant at delivery {RR=5.01 (1.3­19.1) p=0.018} and Traditional Birth Attendant (TBA) delivering the baby {RR=2.7 (1.1­6.4) p=0.03). Effect of sex of the neonate, mother and fathers' ages were not significant at the 5% level in the model. CONCLUSION: Neonatal deaths contribute significantly to the high infant mortality in this rural community. Services provided by TBAs are not optimal but appear to be better than having noone in attendance at delivery. TBAs therefore need to be trained to identify at risk neonates and refer. Obstetric and public health services have to be available and made more accessible at the grass root level