Essential newborn care practices and determinants amongst mothers of infants aged 0 – 6 months in refugee settlements, Adjumani district, west Nile, Uganda

Background Despite recent improvements in child survival, neonatal mortality remains high in most developing countries. Countries affected by humanitarian emergencies continue to report the highest neonatal mortality rates. Objective To assess essential newborn care practices and its determinants amongst mothers of infants aged 0-6 months in refugee settlements in Adjumani district. Methods A cross-sectional study was conducted among mothers of infants aged 0-6 months in refugee settlements, Adjumani district. A total of 561 mothers of infants were selected using systematic sampling technique from households. Data were collected using a semi-structured questionnaire. A composite outcome variable, Essential Newborn Care practices was created by merging different care practices (neonatal feeding, thermal care, and cord care). Multiple logistic regression analysis was used to determine predictors of Essential Newborn Care. Results and conclusions Over half (57%) of the mothers breastfed their newborns within one hour. Half (50.1%) of mothers cleaned the umbilical cord of their newborns. Only 17% of the newborns received optimal thermal care immediately after birth. Mothers aged 20-24 years (OR 0.38, CI 0.17-0.96) and those involved in subsistence farming (OR 0.67, CI 0.38-1.45) were less likely to practice good newborn care compared to those in other occupations. Newborn care practices were sub-optimal in this refugee setting. To improve newborn care practices, there is need to educate mothers through community-based health interventions in order to promote delayed bathing, ideal infant feeding, thermal and umbilical cord care. This was a community based cross-sectional study conducted during October—December 2016 in Adjumani district, west Nile, Uganda. The district has an estimated population of 432,000 inhabitants [26]. These includes an estimated 240,905 refugees living in 17 designated settlements [27]. Health service for refugees and host communities are provided by 42 public health facilities. These include Adjumani district hospital, three Health Centre IV’s and twenty-one Health Centre IIs. Health Centres are part of the district health systems in Uganda. The district health system consists of several levels. Village Health Teams (VHTs) are volunteer community health workers who deliver health education, preventive and simple curative services in communities. The VHTs constitute level one health services. The next are Health Center IIs, that provide outpatient care and are managed by a nurse serving approximately 5,000 people. Health Center IIIs (HC III) serve 10,000 people and provide in addition to HC II services in patient, simple diagnostic and maternal health. It is managed by a clinical officer. Health Center IV’s are managed by a medical doctor and provide surgical services in addition to all other services provided at HC III. Other conditions requiring higher level care are referred and managed at the Adjumani district hospital and Mulago National Referral Hospital. Health facilities provide a range of services including health promotion, curative, Primary Health Care, delivery and minor procedures. Maternal and child health services include ANC, delivery and postnatal care are provided in several facilities. Utilization of these services by mothers in most facilities is high deliveries by skilled health worker at 93% and complete ANC at 84%. Reasons for these high rates of utilization includes close proximity of health facilities in or around settlements and health education and promotion for mother and newborn care in this setting. The sample size was calculated using the Kish and Leslie formulae [28]. A total of 561 mothers was the final sample size for the study. The study population consisted of women of reproductive age between 18–49 years of age who had delivered a baby in the last six months prior to the conduct of this study. The six-month limit was set in order to reduce the chances of recall bias by the mothers of infants. Participants were selected if they were the mother of the infant. Mothers who had lived in the sampled refugee settlements for more than one year were included in the study. Mothers who had a still birth and were seriously ill were excluded. We used the modified WHO, EPI cluster sampling technique [29]. Refugee settlements in the district are divided into zones and blocks. For this study, a zone was designated as a cluster. The sampling interval was determined by dividing the total population by the number of clusters. An identification form was used to identify the clusters that were included in the study. The initial cluster was determined at the centre of each settlement. A total of 11 refugee settlements were purposefully selected for the study. In each zone, a household list was obtained from the local leader (camp commandant). The first respondent in the initial household was randomly selected. The nearest household whose front door was closest to that which was first sampled was visited next. In each homestead, the household head was approached, informed about the study objectives and requested to provide consent. A total of 46 respondents were interviewed in each cluster. In each household, we interviewed one eligible and consenting adult respondent. The respondent was either the mother of the child or a close caretaker. Data were collected using a semi-structured interview questionnaire adapted from the WHO indicators for assessment of Infant and Young Child Feeding (IYCF) [30] and by reviewing other literature. The questionnaire includes items on respondents and household socio-demographic characteristics, antenatal care, birth preparedness, delivery and immediate newborn care, nutrition, postnatal care for mother and baby, neonatal illness and care seeking. The questionnaire has been field tested and used in several settings. The study questionnaire was translated from English into the local language (Juba Arabic) spoken widely by refugees in the study area. This was to make the questions easily understandable to the study population and it was retranslated back into the English language. The survey was conducted from October to December 2016. The questionnaires were administered to respondents by trained research assistants with public health background, skills in data collection and experience from previous studies. The research assistants and supervisors were recruited and trained for two days on the study objectives, data collection procedures and ethics by the authors. The research assistants approached, screened, selected households and administered the questionnaire to participants. The dependent variable was newborn care practices. These included cord care, optimal thermal care and breastfeeding initiation. The dependent variable for the study was a binary outcome created by categorizing three variables including good neonatal feeding that is if breastfeeding was initiated within the first hour after delivery and late initiation if initiated more than 1 hour after birth. This includes the proportion of children aged 0–6 months who were exclusively breastfed, while acknowledging that it may be lower than the current status. Good cord care in this study was defined as use of a clean instrument to cut the umbilical cord, clean thread to tie the cord and no substance applied to the cord. Optimal thermal care referred to placing the baby on Skin-to-Skin contact or wrapping at birth plus giving the first bath after 6 or more hours. Independent variables included mother’s socio-demographic characteristics. Completed questionnaires were checked for completeness and coded. Double data entry was conducted using Epi Data version 3.1. Discrepancies identified in the data were reconciled by referring to the original survey forms. Further data entry was conducted to correct for inconsistencies identified during exploration and analysis and recorded in an analysis log. Data were exported into Stata SE version 14 for analysis. A composite outcome variable was created based on the following variables: (a) Good cord care (b) Thermal care and (c) Good neonatal breastfeeding. The composite variables were dichotomized to Yes (all practices present) or No (one or more practices missing). Descriptive analysis was conducted to produce tables of frequencies and proportions. Binary logistic regression was conducted to determine the association between the independent and dependent predictors of ENC practices. Variables with a p < value less than or equal to 0.05 were considered significant for the multivariable logistic regression analysis. Back ward logistic regression was used with a p < value of less than 0.05 considered significantly associated with ENC practices. The strength of association between dependent and independent variables was expressed by using odds ratio with a 95% confidence interval. This study was approved by the Higher Degrees and Research Ethics Committee of Makerere University School of Public Health and Uganda National Council for Science and Technology. Permission to access the refugee settlements and conduct the study was obtained from the Office of the Prime Minister and United Nations High Commission (UNHCR) and the District Health Office and managers of health facilities in Adjumani district. Written informed consent was obtained from all participants prior to conducting the interviews. Participants were informed that participation in the study was voluntary and they could withdraw at any stage of the interview.