Factors associated with fathers' involvement in infant and young child feeding and nurturing care in the urban slums of Bangladesh: A cross‐sectional study

Abstract Optimal infant and young child feeding (IYCF) and nurturing care during the first years of children's lives are crucial for ensuring their growth and development. The IYCF and nurturing practices are associated with a multifaceted interplay of context‐specific factors where fathers' involvement is necessary. The aim of this study is to explore the current scenario of fathers' involvement in IYCF practices in the urban slums of Bangladesh. A cross‐sectional survey among the residents of the Sat Tola slum in Dhaka, the capital of Bangladesh was conducted. To avoid social desirability bias, 361 mothers of children aged <24 months were interviewed regarding fathers' involvement. An operational definition of father's involvement was developed based on 11 criteria and then scoring was applied to classify ‘Good Involvement’ and associated factors were identified using multivariable logistic regression. Fathers' mean age was around 31 (SD ± 6.09) years and the majority of them (44.32%) completed primary education. Fathers had diversified occupations such as day laborer (32.41%), garment worker (22.71%), and business (14.96%). Factors that were significantly associated with the father's involvement in IYCF practices include educational status (aOR = 3.98, 95% CI = 1.91, 8.26, p < .00) and theiroccupational status (aOR = 0.34, 95% CI = 0.16, 0.70, p = .00). Fathers were more active for their first child (aOR: 1.99, 95% CI = 1.04, 3.79, p = .03). Having child in the age group of 14–20 months (aOR = 2.73, 95% CI = 1.32, 5.64, p = <.01) and wife in the age group of 21–30 years (aOR = 2.34, 95% CI = 1.20, 4.58, p = .01) were significantly associated. The study finding explored that fathers' education and occupation as well as mothers' age and education were significantly associated factors for fathers' involvement in the IYCF practices. Further longitudinal studies are recommended to establish the causal relationship between fathers' involvement with IYCF and their impact on child growth and development.


| INTRODUC TI ON
Ensuring adequate and appropriate nutrition in infancy and early childhood is one of the central prerequisites for the growth and development of children. Poor health among children is significantly associated with the lack of optimum level nutrition during infancy and early childhood. Undernutrition is accountable, directly or indirectly, for half of all under-5 children deaths around the world UNICEF & WHO, 2020). Recent systematic reviews suggest that inappropriate childhood nutrition is also linked with other impairments including stunting, physical disability, childhood obesity, and poor intellectual performance-all of which are growing public health concerns in many countries (Hume-Nixon & Kuper, 2018;Tandon et al., 2016).
To address the problem, the first 2 years of children's life is regarded as a critical window of opportunity since the optimal feeding practice during this time facilitate apposite growth and development (Prentice et al., 2013). Evidence suggests that among the top 15 interventions for preventing under-5 mortality, the two interventions, namely, exclusive breastfeeding up to 6 months of age and introduction to complementary feeding from 6 months, can prevent almost one-fifth of under-five mortality in developing countries (Grantham-McGregor, 2007). Optimal infant and young child feeding (IYCF) is, therefore, regarded as one of the vital strategies for preventive health and nutrition interventions in many global recommendations. The World Health Organization (WHO) made a "Comprehensive implementation plan on maternal, infant and young child nutrition," which has been endorsed in 2012 by the Member States with an aim to promote and support the countries regarding this (McGuire, 2015). In addition to this, there is a separate "Global strategy for infant and young child feeding" uptake by WHO for protecting, promoting, and supporting appropriate IYCF practices across the globe (santé et al., 2003). Despite these policy-level recommendations and initiatives, there are substantial concerns about the nutrition of children in Bangladesh, especially in the urban slum. Although Bangladesh has developed a strong culture of breastfeeding across the country, it has been found from studies that there are inappropriate breastfeeding and complementary feeding practices. According to the latest Bangladesh Urban Health Survey, among the children aged between 6 and 23 months, only one in four of them is receiving the proper IYCF practices in the urban slum areas, which is 40% for nonslums (NIPORT et al., 2015). Moreover, the previous evidences suggest that the rapid urbanization in Bangladesh will result in unprecedented demographic changes as 50% of its population will be living in urban areas by 2050 (Shafique et al., 2018). Most of these urban population growths will be in slums and already 30% of urban populations are living in slum settlements (Adams et al., 2015). Among the urban areas, Dhaka is currently housing 48% of the slum dwellers (Tuthill et al., 2014) which will be expected to be increasing in the future because of the multiple push and pull factors. Therefore, a major portion of infants and young children in the country is living in urban slum areas, which warrants specific attention regarding IYCF and other interventions in order to ensure their optimum nutrition.
In fact, the IYCF practices are associated with multifaceted interplay of context-specific factors such as the educational status of mothers, socioeconomic profile, feeding-related beliefs and preferences at the community level, healthcare access through local healthcare workers, media orientation, and little knowledge of optimal feeding behaviors (Haider et al., 2010;Joshi et al., 2012;Ruel et al., 1999;. Predominantly child health care and nurturing are 'mother centric', and the participation of the father is, thus, not emphasized in most low-income countries (Hackett et al., 2015). However, there are international policy directions that emphasized father's involvement in the IYCF practices. For example, in the International Conference on Population and Development (ICPD), paternal involvement was specifically highlighted for maternal and child care (Tuthill et al., 2014).
The evidence suggests that fathers' involvement has a positive and strong association with children's overall nutrition and cognitive skills development (Fan & Chen, 2001;Jeynes, 2007). Previous studies also recommended that in order to improve the IYCF, fathers should be included in the program design with tailored messages (Sanghvi et al., 2013). The fathers' involvement is also encouraged in current policies and guidelines of Bangladesh. However, in the global literature fathers' involvement in the IYCF practices generally lacks appropriate attention. Similarly, the context of Bangladesh is not well documented too. Specifically, a vast majority of urban poor families are recent migrants from rural areas, with a lack of social capital and thus face livelihood challenges to address health and nutrition problems . Previous studies suggest that fathers' involvement in childcare is largely influenced by the existing gender norms, the father's professional status as the main breadwinner of the family, and mother's engagement in the formal labor market (Raley et al., 2012).
However, there is lack of evidence revealing the realities of fathers' involvement and associated factors such as profession, education, etc. in the urban informal settlement from low-and middle-income Bangladesh, fathers' involvement, gender role, infant and young child feeding, nurturing care, nutrition, slum, urban health countries like Bangladesh. From this perspective, this study aims to explore the current scenario of fathers' involvement in IYCF in urban slums of Bangladesh so that context-specific evidence can be generated regarding the fathers' involvement in order to develop targeted approach for future intervention.

| Study design, sites, and population
A descriptive cross-sectional survey was conducted from September 2020 to February 2021 for collecting data. The study's target population was the dwellers of Sat Tola slum located in the middle of commercial areas named Mohakhali and Tejgaon, at the center of Dhaka, the capital of Bangladesh. The Sat Tola slum was selected as it is one of the largest and most representative slums of Bangladeshi city, where the majority of the residents are involved in economic activities. According to a recent estimate, approximately 50,000 people live in the Sat Tola slum (Lata, 2020).

| Sampling and recruitment of study participants
The sample size for this study was determined using single population proportion formula with the assumptions of 95% confidence level, allowable error (0.05), and assuming the proportion of father's involvement in IYCF practice in the urban slum (p = .5) as there was no similar study in the same setting. Hence, the total required sample size for the study was 384. So, initially, we recruited a total of 384 participants for the study. However, we dropped partially completed interviews and as a result, we found that a total of 361 respondents completed the full interview which was used in the analysis.
We applied systematic sampling technique for household selection. The first household was randomly chosen from the slum located at the approximate geographical center and then the data collectors proceeded to the next household until the required study participants were sampled. From each household, one participant was selected. Considering the objective of the study, we included the women aged 15-49 years having children aged <24 months of age and the women who are usual residents (living for more than 6 months) in the selected slum. Women with speech and hearing disabilities or severe illness, and who did not agree to participate in the study were excluded from the study. Besides, since the study objective is about the fathers' involvement, we excluded those women who are not living with their husbands currently.

| Data collection
We used a structured paper-based questionnaire to collect data from a team consisting of two data collectors. The study investigators (DSB and NA) provided training to the data collectors about the study objectives, methodology, and on the questionnaire. The training sessions also included sessions about rapport building, ethical issues such as confidentiality, cultural awareness, and risk management including COVID-19 preventive measures. Once the training was done, the study investigators arranged a pilot survey, and necessary corrections were made following the pilot.
The data were collected through face-to-face interviews conducted at the respondent's home maintaining privacy. Before the start of the interview, the data collectors clearly communicated that participation in the study was voluntary and no incentive would be given. The study investigators (NA and DSB) supervised the data collection process and reviewed all the data collection sheets for completeness, accuracy, and consistency. From each respondent, data were collected considering both independent and dependent variables of the study. The independent variables included sociodemographic characteristics of the mother, father, and child; as well as the economic condition of the household. Regarding mother's and father's sociodemographic characteristics, we collected their age, education, and occupation information. For the child, we collected child's age, sex, and birth order data. In addition, to understand the economic condition, we collected monthly household income.
For the dependent variable, we asked 11 specific questions to the mother regarding father's involvement in the IYCF practices.
Previous research in a similar country setting (Abera et al., 2017) showed that asking questions to fathers about their involvement in IYCF practices has the potential to bring in the "Social Desirability Bias" since the fathers might have a tendency to report an answer in such a way that they think to get more socially acceptable than what is their "true" answer. In order to avoid the social desirability bias, we directed the questions to mothers only. The questions captured fathers' involvement status in terms of the support they provide to the mothers as well as the direct childcare they do for ensuring the IYCF. Regarding fathers' support, we asked questions on whether the fathers supported mothers on health service utilization, encouraged early initiation of breastfeeding and colostrum feeding, discouraged prelacteal feeding, encouraged breastfeeding on demand, encouraged exclusive breastfeeding for 6 months, and psychologically supported the mothers to ensure appropriate IYCF. We also asked questions on whether the fathers did direct child care, helped mothers by doing other household chores, arranged additional food for mothers, that is, BF mothers, and supported mothers in BF and child feeding in public places.
From these 11 structured questions, we introduced a scoring to identify the Good and Poor Involvement of the fathers. For each "Yes" answer, 1 point was given while 0 was assigned for each "No" answer from the respondent. "Good involvement" was defined if the father scored 6 or more, while "poor involvement" was defined as a score less than 6. These criteria have been selected based on a previous study that explored father's involvement conducted in a similar country setting (Abera et al., 2017). While scoring the father's involvement, we did not separate father's direct childcare issues from father's support to the mother considering previous evidence that father's involvement is a comprehensive construct that includes direct father-child engagement as well as fathers' interaction with other household members especially with mother of the child (Sarkadi et al., 2008).

| Data management and analysis
We analyzed the collected data to present descriptive statistics as percentage (95% CI) for categorical variables or mean ± SD for continuous variables. Bivariate analysis with Chi-square test was performed to see the association between fathers' involvement status and the demographic variables. Finally, we performed a multivariable logistic regression model. In the multivariable model, the variables with a p-value < .25 in the bivariate model were included. The backward elimination process was performed to finalize the multivariate model. We used the dichotomized analysis for this study primarily because of the resource and time constraints.
Besides, the scope of the research question was limited only to see the existing scenario on fathers' involvement in IYCF in urban slums of Bangladesh. We did not explore further how it is associated with child outcomes in terms of socio-emotional development and other family dynamics like motherhood-fatherhood adult adjustment dimension. The statistical analysis was carried out by utilizing the computer software Stata (version 13). Data were checked, edited, and coded properly before performing the analysis.

| Ethical considerations
For the Institutional Review Board (IRB) approval, the ethical approval was taken from North South University (NSU) ethical review committee. Confidentiality was maintained throughout the research process; the interviews were taken in a venue at the convenience of the interviewee where confidentiality was maintained. Besides, the papers containing the information of the study participants remained to the study investigators in a locked cabinet and no one except the investigators involved with this research was able to look at the interview notes or any other information. It was ensured that identifiable information would not be disclosed without the permission of the respondents. Informed consent of the respondents (mothers) was obtained before the interview. Respondents were ensured that they have the right to decline and withdraw from the study at any time would be accepted.
Regarding conflict of interest, the authors disclose no conflict of interest in conducting the study. Informed consent of the respondents (mothers) was obtained before the interview. Respondents were ensured that they have the right to decline and withdraw from the study at any time would be accepted. Regarding conflict of interest, the authors disclose no conflict of interest in conducting the study.

| Demographic characteristics of respondents
A total of 361 mothers who had children aged less than 24 months participated in the study, which gives a response rate of 94.01%, out of 384 mothers attempted. The mean age for the children was 9 (SD ± 6.26) months. Besides, the numbers of male and female children were approximately similar.
The mean age of mothers was found around 25 (SD ± 5.45) years of age. Forty-seven percent (n = 170) mothers got primary education followed by secondary and above education (43%, n = 154).
Regarding occupation, 89.75% of mothers were found homemakers while around 7% were garment workers followed by day laborer and business. Regarding monthly household income, 34.35% (n = 124) reported that their income is within the range of 5000-10,000 BDT ($58-$115; 1USD ~ 87BDT) per month, while 36% (n = 130) was within the range of 11,000-15,000 BDT ($127-$173). Table 1 depicts detailed information on the background characteristics of the study respondents.

| Father's involvement in IYCF
Regarding fathers' involvement, overall, 63% (n = 229) of fathers were found to be categorized as good involvement, while 37% (n = 132) of them were labeled as poorly involved based on the scoring described in Section 2.

| Factors affecting Fathers' involvement in IYCF
Fathers' involvement in IYCF practice was further investigated to find out whether there is any association with socioeconomic demographic variables. Mother's age (p = .049), mothers' education (p < .01), father's education (p < .01), and father's occupation (p < .01) were separately associated with involvement status. Table 2 shows the detailed results of the bivariate analysis.
To observe the adjusted results, we included child's age, child's sex, child's parity, mothers' age, mother's education, father's education, and fathers' occupation in the multivariate model. It was observed that fathers were having about two times higher involvement in IYCF practice for their first child compared to the other children (aOR: 1.99, 95% CI = 1.04, 3.79, p = .03). Besides, those who had 14-20 months aged children were about three times more likely to have good involvement compared to those who had 0-6 months aged children (aOR = 2.73, 95% CI = 1.32,
On the other hand, day laborer fathers are 0.66% less likely (aOR = 0.34, 95% CI=, 0.16, 0.70, p = .00) to have good involvement than the business owner fathers. Table 3 shows the detailed results of multivariate analysis.

| DISCUSS ION
The importance of the involvement of fathers during the complementary feeding period has been underscored across the globe . This research is one of the first studies that provides evidence of the present status of fathers' involvement in urban Bangladesh specifically focusing on the poor living in the slums. The study explored the proportion and associated factors of fathers' involvement in IYCF practices among urban slum dwellers in Bangladesh. It has been observed that the overall 63% of fathers had good involvement in IYCF practices. This is somewhat similar to a study conducted by Abera et al. in Ethiopia in 2017, which reported overall 72.4% involvement of fathers in breastfeeding practices (Abera et al., 2017). The existing literature suggests that fathers play a pivotal role in terms of ensuring child health and survival by providing support and encouragement to the mothers that can strengthen maternal confidence. Especially in terms of breastfeeding, mothers, who are supported by the child's father, were more likely to breastfeed their children for a longer duration (Mannion et al., 2013). Another recent study by Mithra et al. (2021) in India found that almost 41% of fathers had poor involvement in IYCF practices. Considering the geographic condition and sociodemographic profile, this study also has shown similar results where 37% of fathers were observed to have poor involvement in IYCF practice.
The finding suggests that fathers' occupation has a significant association with fathers' involvement in IYCF. This finding corroborates with other similar studies in Bangladesh. One previous research suggests that one of the important factors that is closely associated with the initiation of complementary foods at the recommended age is the occupation of the father . This study revealed that the children whose fathers are engaged in occupations like rickshaw-pulling or small enterprises had an increased risk of not starting complementary feeding in comparison with the fathers engaged in agricultural occupation. Besides, the households with fathers involved in occupations related to agriculture had better food security than the households where fathers are engaged in nonagricultural wage-dependent occupations. This increased level of food security at the household level is linked with better infant feeding practices in Bangladesh . Apart from food security issue, there is also a workplace time dimension of fathers' TA B L E 1 Descriptive statistics of the socioeconomic characteristics of the study participants. .00

TA B L E 3
Multivariate analysis of father's involvement in IYCF practice.
day-laborer fathers have less opportunity of having good involvement in IYCF practice compared to fathers engaged in business. It is plausible that day-laborer fathers need to spend more time in their work, which impedes getting involved in the IYCF or caring practices. A recent systematic review on complementary feeding practices among Bangladeshi infants and young children also suggested that father's occupation is an influencing factor for complementary child-feeding practices (Manikam et al., 2017).
Apart from fathers' occupation, fathers' education was significantly associated with father's involvement in the IYCF practice. It was observed that the more the fathers are educated, there is more chance of fathers' good involvement. A study conducted in Aceh, Indonesia by Ahmad et al. (2018) also has suggested that the father's education is significantly associated with child's nutrition and complementary feeding practices. Besides, the finding of this study is also congruent with the study conducted by The South Asia Infant Feeding Research Network (SAIFRN) which carried out analyses of existing demographic health survey (DHS) data to assess complementary feeding practices across countries in South Asia. The SAIFRN study revealed that low levels of parental education, especially fathers' education, were consistently associated with poor complementary feeding across South Asian countries .
The finding of this study did not reveal significant association between fathers' involvement in IYCF and monthly family income of the household. This finding does not match with other similar studies. A study conducted in Madagascar by Rakotomanana et al. (2017) identified that poor household wealth is associated with poor IYCF practices. The systematic review of Bangladeshi studies revealed that household income is a promoter of IYCF practices (Manikam et al., 2017). However, another study from West Bengal, India reported that although mothers' perception of IYCF was better among higher income households group, lowest practice was observed among this group . However, finding from this study did not get any such association. Besides, the study finding suggests that there was no statistically significant association between fathers' involvement in IYCF and the sex of the child.
This finding is also not aligned with other relevant researches. For example, a study conducted in the coastal region of southern India reported that the fathers of male children were more likely to have poor involvement in IYCF practices compared to those fathers who have female children (Mithra et al., 2021). Similarly, there is speculation from previous studies that a parent's age has the potential to influence his/her time with children. This is because, the younger the parents, the more time they might need to invest in career development, while the older ones might be more secure in their profession allowing them to contribute more to child care (Raley et al., 2012). However, the results of this study did not suggest significant association between father's age and fathers' involvement in IYCF practices.
Regarding child's parity, in this study, first-time fathers have more likelihood to have good involvement in IYCF compared to their experienced counterparts. This finding is similar to another study conducted in India . Furthermore, the study results showed that the fathers of children aged between 14 and 20 months had 2.82 times more chances of good involvement in IYCF practice compared to fathers of children aged 0-6 months of age. This result is congruent with other similar studies that showed fathers take comparatively reduced role in feeding babies at the early stage, but become more responsible in child feeding during the preschool years (Baxter & Smart, 2011;Daniels et al., 2009). A similar study conducted by Tikotzky et al. (2010) in Israel revealed that fathers' involvement was significantly lower in terms of nighttime baby care at an early age.
The limitation of the study is that the emphasis was primarily provided only toward fathers' actual involvement, and details of the other sociodemographic variables like access to media and possession of wealth were not included, which could be an influencing factor in the fathers' involvement in IYCF practices. Furthermore, the study explored the father's involvement from mothers' perspective using a mother-report questionnaire.

| CON CLUS ION
The study finding showed that more than half of the fathers had good involvement in IYCF practice and child care in selected slum areas in Bangladesh. Fathers' education and occupation were the factors that influenced the status of fathers' involvement. Besides, mothers' age and education were also influencing fathers' involvement in the IYCF practices. The child's age was also found toward significant.
However, more than one third of the (35%) fathers had poor involvement. The study results will be helpful to assess the propor-

ACK N OWLED G M ENTS
This study was part of the thesis for the fulfillment of Master's degree in Public Health from North South University, Bangladesh of DSB. DSB is indebted to the Public Health Department of this university and the authors are thankful to the study participants who provided information for this study. icddr,b is grateful to the Governments of Bangladesh, Canada, Sweden, and the UK for providing core/unrestricted support.

FU N D I N G I N FO R M ATI O N
The authors have not received any funding for this work.

CO N FLI C T O F I NTER E S T S TATEM ENT
The authors declare no competing interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.