In Tanzania, suboptimal complementary feeding practices contribute to high stunting rates. Fathers influence complementary feeding practices, and effective strategies are needed to engage them. The objectives of this research were to examine the acceptability and feasibility of (1) tailored complementary feeding recommendations and (2) engaging fathers in complementary feeding. We conducted trials of improved practices with 50 mothers and 40 fathers with children 6–18 months. At visit 1, mothers reported current feeding practices and fathers participated in focus group discussions. At visit 2, mothers and fathers received individual, tailored counselling and chose new practices to try. After 2 weeks, at visit 3, parents were interviewed individually about their experiences. Interview transcripts were analysed thematically. The most frequent feeding issues at visit 1 were the need to thicken porridge, increase dietary diversity, replace sugary snacks and drinks and feed responsively. After counselling, most mothers agreed to try practices to improve diets and fathers agreed to provide informational and instrumental support for complementary feeding, but few agreed to try feeding the child. At follow-up, mothers reported improved child feeding and confirmed fathers’ reports of increased involvement. Most fathers purchased or provided funds for recommended foods; some helped with domestic tasks or fed children. Many participants reported improved spousal communication and cooperation. Families were able to practice recommendations to feed family foods, but high food costs and seasonal unavailability were challenges. It was feasible and acceptable to engage fathers in complementary feeding, but additional strategies are needed to address economic and environmental barriers.
Trials of improved practices (TIPs) is a formative research approach that involves a series of visits to participants’ homes to test and refine nutrition and health recommendations to inform intervention design (Dickin et al., 1997). Based on participants’ current practices, they receive several tailored recommendations and then select which practices to try. TIPs allow participants to try a recommended practice for a period of time and then provide feedback about the acceptability and feasibility based on their experience (Dickin et al., 1997; Dickin & Seim, 2013; Harvey et al., 2013). The results help researchers and program implementers identify potential barriers, develop strategies to overcome those barriers and remove or modify recommendations that are not feasible for participants (Harvey et al., 2013). We conducted three TIPs visits: visit 1 to assess current roles and practices related to complementary feeding, visit 2 to provide tailored recommendations and ask participants to choose practices to try and visit 3 to follow up on participant experiences (Table S1 in the supporting information). Including fathers in TIPs was a novel approach and little was known about how to recommend and motivate specific practices. To better understand norms and attitudes among fathers, we conducted focus group discussions (FGDs) in place of individual interviews with fathers at visit 1. For visit 2, we prepared gender‐specific counselling guides (Supporting Information) based on nutrition counselling materials from the Tanzania Food and Nutrition Centre and previously conducted recipe trials with mothers in the study areas. Mothers who participated in the recipe trials did not participate in TIPs. This study took place in rural areas within three of the five Lake Zone regions (i.e. Geita, Kagera and Mwanza) to represent areas with different staple foods (i.e. banana or maize). The rates of stunting among children under 5 years of age were 39%, 40% and 26% and the proportion of children 6–23 months who consumed a minimum acceptable diet were 28%, 18% and 30%, respectively (Ministry of Health Community Development Gender Elderly and Children (MOHCDGEC) [Tanzania Mainland] et al., 2019). Farming and fishing are common economic activities in the Lake Zone. Sukuma is the predominant ethnic group in Geita and Mwanza, and Haya is the predominant ethnic group in Kagera; both are patrilineal and patrilocal (Manji, 2000; Masele, 2020). In all three regions, it is common for rural families to live together in a homestead with their extended family, including the husband’s parents and siblings (Masele, 2020). In the Lake Zone, as in many settings around the world (Doyle et al., 2014), traditional gender roles are common and social norms hold women responsible for household tasks and caregiving, as confirmed in our previous formative research related to breastfeeding (Matare et al., 2019). Typically, men are not involved in the care and feeding of infants and young children, and women have little decision‐making power (Masele, 2020). Women’s empowerment indicators are lower in the Lake Zone compared to national levels (MoHCDGEC, 2015). Only 25% of married women in the Lake Zone reported making decisions either by themselves or jointly with their husband about their own health, major household purchases, and visiting their family, compared with 35% nationally. Similarly, 68% of women of reproductive age in the Lake Zone agree that a husband is justified in hitting or beating his wife for specific reasons, compared with 58% nationally (MoHCDGEC, 2015). Two wards in each region were purposively selected to ensure diversity in population size and access to the district town centre, healthcare facilities and lake. One village was randomly selected from each ward. Based on accepted sampling approaches for qualitative research (Hagaman & Wutich, 2017; Sobal, 2001), the target sample size in each village was nine couples, or 54 total. Community health workers helped identify potential participants with infants and young children 6–18 months of age using predetermined criteria to ensure variation in child’s age and mother’s age, parity and educational status. Exclusion criteria were maternal age less than 15 years, a health condition that would interfere with trying recommended behaviours (either mother or child) and travel plans that would preclude follow‐up visits. After mothers consented to participate in the study, their male partners were invited to participate. One FGD was conducted in each village with fathers participating in TIPs. Additional fathers of children 6–18 months of age were recruited to participate in FGDs to have at least 10 participants per group. Three female and three male interviewers conducted interviews in Swahili. All interviewers were Tanzanian and were experienced qualitative data collectors, with bachelor’s or master’s degrees in nutrition or public health. They received a 10‐day training in TIPs and complementary feeding and were supervised by a community development specialist and two nutritionists. Interviewers were trained to share recommendations, but not pressure participants to select them as we were interested in participants’ responses to the recommendations. Each participant was interviewed by the same interviewer at each visit. Interview‐interviewee pairs were gender‐concordant. With participant consent, all interviews and FGDs were audio recorded. Interviewers collected data from mothers and fathers at three time points (Table S1). For mothers, at visit 1 they were asked about current feeding practices, which included infant feeding practices in the previous 24‐h period (World Health Organization, 2010) and 7‐day dietary diversity assessment to capture foods not eaten every day developed for this study; household hunger (Ballard et al., 2011); support they receive for complementary feeding; and additional support they would like. Fathers participated in FGDs for visit 1, followed by two visits. Focus group discussions with fathers explored knowledge and beliefs about child feeding, gender norms and roles related to the care and feeding of young children, and potential ways fathers could become more involved in complementary feeding. As part of the FGDs with fathers, we used a vignette (i.e. short story without an ending) to encourage reflection on the types of support husbands could provide for complementary feeding, a method that has been used successfully in Tanzania (Gourlay et al., 2014). The vignette described two brothers with different levels of involvement with their young children, and participants were encouraged to reflect on the circumstances in the story, suggest actions characters should take, and imagine how the story would end. After the mother’s first visit, the female interviewer shared mothers’ current feeding practices and concerns with one to two supervisors and the male interviewer that would interview the father. As a group, they used the counselling guides to discuss potential recommendations and decided which recommendations to offer mothers and fathers at visit 2 based on feeding practices and family context. Families received different recommendations based on their circumstances. While the topics in the counselling guide were similar for mothers and fathers (e.g. feeding animal‐source foods, replacing sugary snacks with healthy ones), the specific actions were different based on their roles. Recommendations for mothers focused on direct feeding (e.g. feed animal‐source foods), whereas those for fathers included support for complementary feeding (e.g. purchase animal‐source foods) as well as direct feeding practices. At visit 2, mothers and fathers were counselled individually about ways to improve complementary feeding practices. Interviewers asked participants to select one to two practices to try over a 2‐week period. Mothers and fathers were counselled individually to allow participants to speak freely about their thoughts and experiences about complementary feeding recommendations and father involvement. At visit 3, two weeks after receiving tailored nutrition counselling and selecting new practices to try, individual in‐depth interviews with mothers and fathers explored their experiences. Participants’ selection and trial of recommended practices were tracked and tallied. Tallying allows comparison across recommendations to identify which were most commonly recommended to participants (reflecting frequency of non‐ideal practices) and which were most acceptable (reflecting initial perceptions of acceptability) (Dickin & Seim, 2013). We used Microsoft Excel (Microsoft, Redmond, WA) to summarise sociodemographic characteristics, child dietary recall data and participants’ behavioural choices. Qualitative data included field notes from mothers’ visits 1 and 2 and fathers’ visit 2, as well as verbatim transcripts of the fathers’ visit 1 FGDs and visit 3 in‐depth interviews for mothers and fathers. Interviewers took detailed notes during visits 1 and 2 and expanded their notes after each interview. Field notes were translated into English by independent translators and reviewed by each interviewer for accuracy. An independent translation firm transcribed father FGDs and mothers and fathers visit 3 interviews verbatim in Swahili, and then translated them into English. Father FGDs were coded and analysed thematically by three study team members. Key themes were identified and summarised in a matrix. English transcripts of visit 3 interviews with mothers and fathers were qualitatively analysed by seven team members. First, we manually coded the women’s transcripts. To ensure coding consistency, all analysts independently read and manually coded the same two transcripts. The team then reviewed the coded transcripts as a group and collaboratively created a codebook. The team agreed on code names, definitions and inclusion and exclusion criteria (DeCuir‐Gunby et al., 2010; Macqueen et al., 1996). We continued to individually code and review transcripts with frequent debriefings and codebook refinement. After reviewing the same five transcripts, we determined that the level of coding agreement was consistent across the analysis team, and no longer required each transcript be read by all team members. All transcripts were then coded in Atlas.ti Version 8 (Scientific Software Development, Berlin, Germany) with at least two analysts independently coding each transcript. The analysis team met frequently to discuss emergent codes and any discrepancies until reaching consensus, iteratively revising the codebook. Major themes and findings were discussed during peer debriefings. The analysis team followed the same approach for the fathers’ visit 3 transcripts, starting with the same codebook and adding additional codes specific to fathers’ experiences. In addition, visit 3 interview transcripts from mother–father pairs were analysed together as a unit and summarised in a matrix to examine couple communication and cooperation and consistency of responses within couples. The Cornell University Institutional Review Board and the National Institute for Medical Research in Tanzania approved this study. All participants gave written informed consent.
DIMA AI Care
