Menu
Menu
Menu
Menu
Appetite in children is an important determinant of nutritional intake and growth. The information used by caregivers to understand children’s appetite can help inform infant and young child feeding promotion and appetite assessment. We conducted a qualitative study to (a) explore maternal perceptions and responses to children’s appetite and (b) to identify how these factors differ by type of caregiver, level of maternal experience, and urban versus rural context. We used purposive sampling to recruit mothers and alternate caregivers into 14 total focus group discussions (six to eight participants in each group; N = 95) in both urban and rural settings in Bangladesh. To understand children’s appetite, caregivers monitor children’s dietary patterns, emotional signs, and physical and verbal cues. Healthy appetite was observed by willingness to eat diverse foods, finish offered portions, and by acceptance of foods without excessive prompting. Child illness was cited for a cause of low appetite, which was manifested through fussiness, and avoiding commonly consumed foods. Mothers described a limited set of feeding practices (offering diverse foods, playing, and cheering children with videos) to encourage consumption when children lacked appetite. Mothers’ stress related to work was noted as a barrier to identifying appetite cues. Urban mothers described a lower access to instrumental social support for child feeding but informational support than mothers in the rural setting. Understanding caregivers’ perceptions of children’s appetite may inform strategies to improve responsive feeding and tool development to assess changes in appetite as early indicators of change in health or nutrition status among high-risk children.
We conducted 14 focus group discussions (FGDs), six in a rural setting, and eight in an urban setting in Bangladesh. IYCF practices in Bangladesh have improved steadily in the past 10 years but remain poor. Nationally, only 23% of children 6–24 months are fed a “minimally acceptable diet”—a composite indicator of child feeding adequacy that includes the frequency of feeding and dietary diversity (NIPORT, Mitra, & Associates, and IRC Macro, 2016). Both urban and rural areas were chosen for this study to broaden the generalizability of the study findings. The urban setting, Mirpur, is located in Dhaka and has a population of approximately 500,000 residents in an area of 14 km2 (35,000/km2; Ahmed et al., 2014). This community was selected as the urban setting because it is inhabited by low‐ and middle‐income families, and the residential and sanitary conditions are typical of a congested urban slum settlement. The study investigators also have ongoing research activities in this area, and the urban FGDs were conducted in a preexisting feeding centre. The rural site, Mirzapur, is located approximately 75 km from Dhaka. It is a subdistrict located in the Tangail District of Dhaka Division and has a relatively lower population density of 1,091 residents/km2. Approximately half of the households use improved sanitation, 60% of the population has electricity, and the residents rely mainly on tube wells for drinking water. Men are mostly engaged in rice and jute production or daily wage labour, often abroad, whereas women work mainly in the home (Das et al., 2013). The rural FGDs were conducted in a field office of the International Centre for Diarrheal Diseases, Bangladesh (icddr,b). We adhered to the Consolidated Criteria for Reporting Qualitative Research (Tong, Sainsbury, & Craig, 2007). We recruited mothers and alternate caregivers of children between 6 and 59 months to ensure that most children were receiving complementary foods. Data collection occurred in two phases. The first data collection phase recruited FGD participants into groups on the basis of characteristics that the study team believed to be important factors for child appetite. In this phase, we conducted 12 total groups, six in the urban setting and six in the rural setting. Within each setting, we further broke groups down by the age of the child and caregiver experience. Within each site, child age (6–12 months, 13–24 months, and 25–59 months) was used as a group characteristic to account for the different developmental stages of children in relation to how appetite was demonstrated to caregivers. To investigate differences by caregiver experience, two focus groups in each site were composed: one consisting of experienced mothers—that is, those who had more than one child—and one with first‐time mothers. To explore whether mothers’ employment status has any effect on understanding or recognizing their children’s appetite, we conducted two additional focus groups in the urban setting among mothers who were employed outside of the home. We did not conduct a FGD of this same group in the rural setting as mothers in this context are mostly work full‐time at home are caregivers. The breakdown of focus groups and inclusion criteria for each is summarized in Table 1. Though mothers are still thought to be the primary arbitrator in determining feeding responsibilities (Troiano, Briefel, Carroll, & Bialostosky, 2000), changing maternal employment trends have led to an increasing involvement of alternate caregivers (Chao & Rones, 2007). In Bangladesh, 32% of ever‐married women are employed (NIPORT, Mitra, & Associates, and IRC Macro, 2016). Finally, one group in each site was composed of alternate caregivers, which included fathers, aunts, and grandmothers. This group was recruited by screening households through door‐to‐door visits within each study setting. To be eligible, alternative caregivers were required to be adult who provided full or part‐time child care to a child ages 6 to 59 months who was eating foods in additional to breast milk. Fathers and paternal grandmothers (i.e., mother‐in‐laws to children’s mothers) were specifically recruited because they are known to be key decision‐makers for child health in this context (Shahabuddin et al., 2016). Each focus group consisted of six to eight mothers and/or alternate caregivers (N = 95). The data collection was conducted between June 2016 and February 2017. Summary of focus group sampling strategy and recruitment The conceptual framework of the study (Figure 1) was used to inform the research question, develop the FGD guide, and to guide the analysis. The research team created this framework early in the research process to inform the sampling plan and questionnaire development. This framework describes the proposed basic, underlying, and immediate influences of caregiver’s recognition of children’s appetite and is based on the WHO Conceptual Framework on Childhood Stunting, which focuses on community, societal, household and family influences of inadequate complementary feeding practices (Stewart et al., 2013). Examples of immediate influences include whether grandmothers, aunts, or hired caretakers are the main participants in child feeding, the number of previous children cared for by a mother, and the level of instruction a mother has received on treating or feeding sick children. Underlying influences include the work and time‐related constraints imposed by work and other caring responsibilities, mothers’ experiences in urban or rural settings, and norms around feeding (i.e., the timing of the introduction of complementary foods, typical child feeding styles). Basic influences include women’s empowerment, opportunity structure in their surrounding society, educational attainment, and agency. Empowerment is defined as the process by which mothers gain control over the factors and decisions that shape their lives (World Health Organization, 1998). Agency refers to the capacity of mothers to act in their given contexts (Crocker & Robeyns, 2009). Collectively, these factors are expected to exert an influence of mothers’ knowledge, attitudes, and beliefs about child appetite, which comprise their perceptions of appetite. Examples of caregiver time include maternity leave policies, commuting time, and type of work. The geographic context is broadly defined in Bangladesh as urban or rural. The household context and family structure refers to the size of the household and presence of alternate caregivers. The cultural norms of child care and feeding include factors such as normative breastfeeding and complementary feeding practices and styles, and caregivers concern over growth and thinness. Conceptual framework of multi‐level influences of caregiver perceptions of child appetite in low‐ and middle‐income countries The research team has a longstanding relationship with the study communities and maintains a listing of households that include demographic information. Using this register, we identified households in Mirpur and Mirzapur that contained participants who fit the eligibility criteria. Field staff visited eligible participants at home and explained the study objectives. Potential participants were asked if they would be willing to attend a FGD about caregiver experiences with understanding appetite in young children. Eligibility criteria for the study were (a) having a child 24 months and under (for mothers); (b) providing substantial child care for a child under 24 months (for female alternate caregivers) or 5 years (for fathers); and (c) being a full‐time resident in the community of the study setting. We used a higher child age eligibility for fathers to increase the likelihood of recruitment of this group. For the FGD among working mothers, participants were required to be engaged in formal employment outside of the home for at least 20 hr per week for at least 2 months preceding the date of the interview. The day, time, and location of the FGD were confirmed 1 day before the discussion. Participants were briefed about the benefits, privacy, and confidentiality of the study. After obtaining written informed consent and separate permission to audio record the interview from each participant, a research assistant administered a verbal demographic survey with each participant. FDGs lasted approximately 90 min and were all led by the same moderator in Bangla. The focus group guide (Supplemental Table 1) focused on the key topics of interest: (a) words and phrases used to describe appetite; (b) physical and emotional cues/signs/symptoms used to describe appetite; and (c) maternal factors that influence awareness of cues. Participants were provided compensation to cover travel costs and wage loss incurred during participation in the FGD. All study procedures were approved by the International Centre for Diarrheal Diseases Research, Bangladesh (icddr,b) Institutional Review Board. The University of Washington Institutional Review Board provided a waiver, as affiliated researchers only interacted with de‐identified data. Interviews were audio recorded, transcribed verbatim in Bangla, and then translated into English. Participants were noted in interview transcripts as P1, P2, and so forth. Interview transcripts were uploaded into Dedoose qualitative data analysis software (Dedoose Version 7.0.23, 2016;) and double coded by two trained qualitative researchers (redacted for review) who developed the codebook with the principal investigator of the study. Agreement of any discordant codes was achieved through discussion, with any final arbitration by the main analyst (SI) as needed. Direct quotes were extracted from interviews and linked to the demographic data using the de‐identified participant identification number. Codes were developed from the conceptual framework (Figure 1) and through emergent themes. The different participant groups were analysed using the constant comparative method using both explicit and implicit processes (Glaser, 2008). Explicit comparisons were made when one participant group remarked on their identity in that group (e.g., first‐time mothers and working mothers). Implicit comparisons were analysed by comparing main themes between groups (e.g., urban and rural) to identify differences in main findings, including the absence of patterns. For example, if participants in the rural group did not describe the influence of maternal employment on appetite recognition when that topic was assessed, we understood this topic to be less influential on mothers’ appetite recognition in this setting, compared to the urban context where it was discussed in depth in multiple groups.
N/A
