Menu
Menu
Menu
Menu
Background: Recommendations for care in the first week of a newborn’s life include thermal care practices such as drying and wrapping, skin to skin contact, immediate breastfeeding and delayed bathing. This paper examines beliefs and practices related to neonatal thermal care in three African countries. Methods: Data were collected in the same way in each site and included 16-20 narrative interviews with recent mothers, eight observations of neonatal bathing, and in-depth interviews with 12-16 mothers, 9-12 grandmothers, eight health workers and 0-12 birth attendants in each site. Results: We found similarities across sites in relation to understanding the importance of warmth, a lack of opportunities for skin to skin care, beliefs about the importance of several baths per day and beliefs that the Vernix caseosa was related to poor maternal behaviours. There was variation between sites in beliefs and practices around wrapping and drying after delivery, and the timing of the first bath with recent behavior change in some sites. There was near universal early bathing of babies in both Nigerian sites. This was linked to a deep-rooted belief about body odour. When asked about keeping the baby warm, respondents across the sites rarely mentioned recommended thermal care practices, suggesting that these are not perceived as salient. Conclusion: More effort is needed to promote appropriate thermal care practices both in facilities and at home. Programmers should be aware that changing deep rooted practices, such as early bathing in Nigeria, may take time and should utilize the current beliefs in the importance of neonatal warmth to facilitate behaviour change.
We collected qualitative data on thermal care beliefs practices from one Local Government Area (LGA) in Ekiti State in Southwest Nigeria and two LGAs in Borno State in North East Nigeria, two districts in the Oromiya region of Ethiopia and four districts in Lindi and Mtwara regions of Tanzania. These sites were selected because of their high neonatal mortality burden, and were diverse in terms of literacy levels, infrastructure, and health care utilization (Table 1). Within study sites, four typical communities were selected to reflect study site diversity in characteristics that could influence newborn care practices such as access to health facilities, ethnicity and geography. In Tanzania, a newborn care trial was being conducted in the study area [13], so data collection was limited to the control areas of this trial. Data were collected during the rainy/cooler season in all sites. Study site characteristics Data collection included newborn care narratives, observations of bathing and in-depth interviews (IDIs) with recent mothers, grandmothers, fathers, health workers and birth attendants. Data were not collected from birth attendants in Ethiopia as they were rarely used in the study site. The use of multiple methods and a wide range of respondents allowed us to understand thermal care from different perspectives and to corroborate findings. Data were collected as part of a study exploring the potential for emollient therapy in African settings and included specific questions on thermal care. The newborn care narratives collected data on personal experiences and allowed us to understand how events influenced each other. The in-depth interviews collected data on normative behaviors and on the respondents’ experience and beliefs around thermal care practices. The bathing observations aimed to provide a deeper understanding of how practices were actually done and included measuring the length of time the newborn was undressed. Sample size was based on the concept of saturation sampling, with data collection ending when no new information emerged. This resulted in slightly different sample sizes per site with 16–20 newborn care narratives, eight observations, 12–16 mother IDIs, 9–12 grandmother IDIs, eight health worker IDIs and 0–12 birth attendant IDIs. Community informants identified respondents by word of mouth, or snowball sampling. Mothers for the narrative and IDIs were purposively sampled to ensure a range of maternal ages, parities and sex of child and, where these varied, place of delivery, education level, socio economic status, ethnicity and religion. The characteristics of the narrative women are shown in Table 2, no one refused to participate. Characteristics of the women completing narrative interviews Data were collected between July and November 2011 and data collection was guided by a study protocol; interview guides were developed by the research team and adapted for each site through pre-testing. Data were collected in the local language by 3–4 trained interviewers in each site. Interviews were conducted in the respondents’ home or workplace and lasted between 30 and 90 min. All interviews were tape-recorded and field notes taken, and these were used to write expanded notes in Microsoft Word, which included verbatim quotes and interviewer observations and reflections [14]. Bathing observations consisted of one person videoing the practice and another taking notes and asking clarifying questions at the end of the observation. Written consent was gained from all participants and ethical clearance was obtained from University College London Research Ethics Committee, Obafemi Awolowo University Teaching Hospital Ethical Review Board, Ekiti State Ministry of Health Review Board, the Research Ethical Review Committee of the Oromia Regional Health Bureau, the University of Maiduguri Teaching Hospital Ethical Committee and Ifakara Health Institute Institutional Review Board. The site and study coordinators reviewed the expanded notes and tape recordings, and interviewers were provided with feedback on their probing and expanded notes. Regular team meetings were held which included self-reflection and a discussion of methodological issues and emerging themes. Half way through data collection, teams documented key themes in a matrix and modified the guides to ensure missing areas were filled and to remove questions for which saturation had been reached. The study coordinator attended all the training sessions and visited each site during data collection to ensure that comparable methods were being used across the sites. Formal analysis started with re-reading the transcripts to ensure familiarization. This was followed by group coding of 2–3 interviews to enhance conceptual thinking and rigour [15], and individual coding of the same interview to encourage standardized coding. This initial coding, along with the matrix completed during data collection, was used to develop a codebook and a coding template in NVivo. Sites then coded all interviews using the NVivo template, adding new codes and themes as they emerged. The data were then categorized, organized and interpreted. The NVivo files were sent to the Principal Investigator, who re-coded a sub-set of transcripts and compared and discussed codes with the team. In addition to coding in NVivo, a framework approach using Microsoft Excel was used for the narratives so that themes could be more easily compared and contrasted across and within cases [16]. The video observations were used to provide insight into how practices were performed and to determine the length of exposure during bathing and related activities.
N/A
