Women’s autonomy and men’s involvement in child care and feeding as predictors of infant and young child anthropometric indices in coffee farming households of Jimma Zone, South West of Ethiopia

Background Most of child mortality and under nutrition in developing world were attributed to suboptimal childcare and feeding, which needs detailed investigation beyond the proximal factors. This study was conducted with the aim of assessing associations of women’s autonomy and men’s involvement with child anthropometric indices in cash crop livelihood areas of South West Ethiopia. Methods Multi-stage stratified sampling was used to select 749 farming households living in three coffee producing sub-districts of Jimma zone, Ethiopia. Domains of women’s Autonomy were measured by a tool adapted from demographic health survey. A model for determination of paternal involvement in childcare was employed. Caring practices were assessed through the WHO Infant and young child feeding practice core indicators. Length and weight measurements were taken in duplicate using standard techniques. Data were analyzed using SPSS for windows version 21. A multivariable linear regression was used to predict weight for height Z-scores and length for age Z-scores after adjusting for various factors. Results The mean (sd) scores of weight for age (WAZ), height for age (HAZ), weight for height (WHZ) and BMI for age (BAZ) was -0.52(1.26), -0.73(1.43), -0.13(1.34) and -0.1(1.39) respectively. The results of multi variable linear regression analyses showed that WHZ scores of children of mothers who had autonomy of conducting big purchase were higher by 0.42 compared to children’s whose mothers had not. In addition, a child whose father was involved in childcare and feeding had higher HAZ score by 0.1. Regarding age, as for every month increase in age of child, a 0.04 point decrease in HAZ score and a 0.01 point decrease in WHZ were noted. Similarly, a child living in food insecure households had lower HAZ score by 0.29 compared to child of food secured households. As family size increased by a person a WHZ score of a child is decreased by 0.08. WHZ and HAZ scores of male child was found lower by 0.25 and 0.38 respectively compared to a female child of same age. Conclusion Women’s autonomy and men’s involvement appeared in tandem with better child anthropometric outcomes. Nutrition interventions in such setting should integrate enhancing women’s autonomy over resource and men’s involvement in childcare and feeding, in addition to food security measures. A community based cross-sectional study was conducted on Infant and young Childs of coffee producing households of Jimma Zone, Southwest Ethiopia. Jimma zone is one of the 18 zones of Oromia region which is believed to be the birthplace of Coffee [43]. Organic coffee of Jimma zone is the backbone of foreign exchange of the country, which accounts for 4.2 percent of the total world coffee production, sustaining 15 million Ethiopians in its economic chain [44]. According to 2007 national census, the total population and households of the zone were 2,495,795and 521,506 respectively. This zone covers a total area of 15,569 Km2, with reliable rain fall ranging from 1,200–2,800 mm per annum [45–46]. The Sample size for the study was calculated using a prevalence of malnutrition in Mana Woreda of Jimma Zone (42%), a design effect of 2 and a margin of error of 0.05 [47]. A total sample size of 749 was estimated to have a power of 80, calculated using Epi info Version 7 open source sample size calculator. The inclusion criteria was being an infant or young child of permanently registered resident farming household of the Woredas, while exclusions were made on children with severe acute malnutrition warranting referral to nutrition rehabilitation program, severe illness with clinical complications warranting hospital referral and presence of obvious congenital or chronic abnormalities that impair feeding or physical growth measurements. Multi-stage stratified sampling was used to collect data from respondents across the zone. First, three of the nine top coffee producing Woredas of Jimma zone (Mana, Gomma and Limukossa) were randomly selected. Then, the Woredaswere stratified by urban and rural areas of residence and finally one third of villages (Gots) in rural areas and kebeles in urban setting were selected and used as primary sampling unit, followed by a random selection of households with young infant and young child using health extension workers’ registry as a frame. The sample size in each stage was allocated based on proportional to size allocation methods based on central statistics agency report of 2007 [45]. In the event where more than one eligible child was found in a house, the youngest was taken. A structured questionnaire was used for face to face interview of mothers/caregivers. The two immediate causes of malnutrition, inadequate dietary intake and diseases were assessed by dietary methods and morbidity reports respectively. Exclusive breast feeding under the age of 6 months and dietary diversity with feeding frequency for 6–24 months of age children were used as a proxy measure of optimal feeding. The three underlying causes of malnutrition food access, hygiene and childcare were assessed using household food insecurity scale (HFIAS), diarrheal morbidity report (as a proxy indicator of hygiene)and the WHO Infant and young child feeding (IYCF) practice core indicators respectively. All values to indicate optimal practice were based on age specific guideline of WHO and their compliance is considered as optimal childcare and feeding [8–11]. The basic factor for optimal nutrition is assessed by collecting data on socio-demographic variables, households’ assets and utilities, maternal, paternal and child characteristics. The interview were made by trained nurses while anthropometric measurements were taken by three trained graduate nutritionist. Ethical clearance was obtained from the institutional review board of collage of Health sciences, Jimma University, Ethiopia. Respective government and health institutions and local administrators were requested for permission of entry using an official letter from the university. Detailed description of the study to Kebele and “Got” leaders and households were provided while separate informed verbal and written consent for each study participant were obtained. Women’s autonomy was measured by four theoretical proxy domains adapted from DHS tool; ‘mobility’, ‘decision regarding child’, ‘decision regarding family planning’ and ‘finance’. We inquired the mother eight items with binary ‘yes’ or ‘no’ answers, where, ‘0’ represented a no autonomy or involvement and ‘1’ represented a higher level. The first three questions were related to ‘mobility’, asking the mother if she required approval from her husband or family member to go to ‘outside home’, or ‘market place’, or ‘health institution’. The next three questions were related to ‘mother involvement in decision making regarding her child’; specifically, ‘when child got sick’, or ‘child schooling’ or ‘to whom to marry’. The third group of questions related to ‘financial autonomy’ inquiring mothers autonomy on ‘purchase of food’ or ‘big item such as oxen, land and house’. We also asked a single item on autonomy of ‘utilization of family planning service’. Similarly, Paternal involvement in childcare is assessed by five theoretical proxy domains drawn and adapted from Lamb et al., (1987); ‘presence’, ‘engagement in care’, ‘finance’, ‘child health care seeking’ and ‘informational role’[48].Among the above paternal involvement variables ‘paternal engagement in care’ was assessed by two questions. The first item inquired whether the father had ‘engaged in feeding’ of his child. The second question probed the father ‘engagement in child hygiene and psychosocial support’ such as diapering, bathing, handling and playing. Affirmative responses for both questions were set as criterion for a father with a child of 6–24 months of age for ‘optimal paternal involvement in childcare ‘.For those fathers with a child 0–6 month’s age, only the second question was taken as a criterion. ‘Paternal presence’ was determined by calculating the ratio of months at which the ‘father lived with the child in the same roof’ to ‘the child age’. ‘Paternal involvement in child health care seeking’ was assessed by asking the father ‘if he ever brought his child to health institutions since his birth’. Meeting “informational role of the father “was assessed by asking the mother ‘if she had ever received information about optimal childcare from the father of her child or not’. Household Food Insecurity Access Scale (HFIAS) version 3 was used to measure household food security status. HFIAS has been developed by FAO and Food and Nutrition Technical Assistance (FANTA) and validated for use in Ethiopia [49]. Though adaptation for local context is highly recommended in different studies, we used the tool as it is (without change) for the benefit of its ascertained validity and reliability in Ethiopia [50–51]. The instrument has nine items categorized in three domains, anxiety and uncertainty, Insufficient Quality and insufficient food intake and its physical consequences. Definitions of the HFIAS instrument were used to label households as food secure or insecure [49]. Dietary diversity of children was measured using FANTA tool as recommended by the WHO Infant and young child feeding (IYCF) recommendations guideline [8–10]. Optimal achievement of minimum dietary diversity was defined as proportion of children with 6–23 months of age who received foods from four or more food groups of the seven food groups. The seven foods groups used for tabulation of this indicator were adapted for local food items. For example we added “Teff” a local cereal in the grains list of the probing instrument. Consumption of any amount from each food group was sufficient to ‘count’, i.e., there was no minimum quantity, except if an item was only used as a condiment (S1 File). In the same manner, we adopted the WHO IYCF feeding recommendation definitions to assess children’s achievements for Minimum meal frequency [8–10]. Accordingly, the Minimum frequency was defined as proportion of breastfed and non-breastfed children aged 6–23 months who received solid, semisolid, or soft foods twice for breastfed infants 6–8 months, three times for breastfed children 9–23 months, and four times for non-breastfed children 6–23 months. Length and weight measurements were taken in duplicate using calibrated equipment and standardized techniques. Length [height] was measured in the recumbent position to the nearest 0.1 cm using a measuring board with an upright base and movable headpiece made by Seca, Germany. Weight was measured using weighing scales (Seca, Germany) (+10 g precision) with light clothing. Data were entered into EpiData to control skip patterns and allow double entry and exported to SPSS version 21 for analysis. Anthropometric data were analyzed using WHO Anthro version 3.2.2. In the analysis, plausibility of anthropometric Z scores were checked using the WHO protocol recommendations (2006), which provide standard deviation cut points for anthropometric Z-scores as a data quality assessment tool [52].Accordingly, implausible z scores data were excluded if a child’s HAZ was below –6 or above +6, WAZ below –6 or above +5, WHZ below –5 or above +5, or BMIZ below –5 or above +5. Wealth index was generated using Principal Components Analysis (PCA).The scores for 25 types of assets and utilities were translated into latent factors and the first factor that explained most of the variation was used to group study households into wealth tertile. Each question on domains of autonomy and men’s involvement were summed up for their category to generate count base index. Under nutrition were defined based on their indices including: weight-for-age Z-score (WAZ), height-for-age Z-score (HAZ), weight-for-height Z-score (WHZ) and BMI for age Z-score (BAZ). The World Health Organization Child Growth Standards were used to classify nutritional status [53]. Accordingly, children whose weight-for-age z scores was less than -2 SDs below the median for their age and gender were defined as being underweight. Children with height-for-age z scores less than -2 SDs below the median were defined as being stunted and those with weight-for-height Z scores less than -2 SDs below the median was considered as wasted. Severe anthropometric failure is also defined as less than -3 SDs below the World Health Organization determined median scores for each indexes. A multiple linear regression was conducted to isolate independent predictors of nutritional outcomes of child using SPSS version 21 windows software.