A qualitative study of the infant feeding beliefs and behaviours of mothers with low educational attainment

A qualitative study design was adopted to explore perceptions and behaviours of mothers about infant feeding practices by conducting one-on-one telephone interviews using a semi-structured interview guide. This approach was selected not only because it is an effective means of qualitative enquiry [46], but also because it allowed flexibility in interview times and locations, which we deemed essential in being able to reach a range of mothers with infants. Although telephone interviews have the disadvantage that visual cues (e.g., body language and facial expressions) are absent [47] it provides advantages of greater participant anonymity and cost effectiveness [47, 48]. Ethics approval was granted by University of Technology Sydney Human Research Ethics Committee (2013000463).

Participants were recruited from two Australian regions (Australian Capital Territory, ACT and New South Wales, NSW). NSW and the ACT contain approximately a third of the total Australian population in both rural and urban settings. Mothers were eligible to participate if they had not completed a university degree (considered low educational attainment [49]; were the primary caregiver, were fluent in English, had an infant with no major health problems that may affect feeding, eating or growth (e.g., failure to thrive, chronic illness). The participants’ education level was used as the indicator of socioeconomic disadvantage as it has been shown, relative to other commonly used proxies for SEP (e.g., income or occupation) to be most strongly associated with the related concept of maternal diet [50] and has previously been used in our team’s research on feeding practices [51]. We targeted mothers with infants aged up to twelve months to allow us to capture the range of beliefs and behaviours associated with various infant feeding milestones.

The study was advertised in the Playgroups NSW e-newsletter between January and March 2014. This newsletter is sent once a month to the 25,000 members of Playgroups NSW, a free program for parents and carers with children aged 0–5 years. Mothers who saw the advertisement in the newsletter subsequently shared the survey link with other mothers via social media, including a large Facebook group of mothers living in the ACT. The advertisement included a link to a web-based survey (Survey monkey®) where the interested mother provided demographic and contact details. These mothers were then screened according to their education level and age of their infant to assess their eligibility. Eligible participants were then sent a plain language participant letter and a consent form via e-mail. Mothers were asked to verbally consent to the study at the time of the interview and therefore no written or electronic consent form was completed.

The semi-structured interview guide was developed and structured in a way to enable us to address each of Michie et al’s COM-B framework components (Table 1). That is, we designed questions to explore the conditions that may affect each of the target behaviours. The target behaviours and their competing alternatives were informed by the literature as key behaviours related to obesity prevention in early life and included (a) initiating breastfeeding/formula feeding, (b) prolonging breastfeeding/replacing breast milk with formula, (c) introducing solids earlier than four months of age/delaying the introduction of solid foods until around 6 months of age, (d) feeding to appetite/use of non-nutritive (i.e., feeding for reasons other than hunger) feeding, and (e) introducing healthy first foods/introducing unhealthy first foods.

Interview guides were adapted according to the age and feeding milestones of the infant. For instance, mothers who had not yet introduced solids to their infants were not asked questions about their current solid food feeding behaviours but rather their intentions to introduce solid foods. The interview was piloted with 5 mothers meeting the same eligibility criteria as the main study. Refinements were made to the interview schedule to improve clarity and flow. Mothers were interviewed over the telephone by two of the investigators (ST and LA) at a time convenient to them. Interviews were audio recorded with participants’ permission. Mothers were offered an AUD30 supermarket voucher in appreciation of their time.

Interviews were transcribed verbatim and five randomly selected interviews were checked against the interview recording by ST to assess the accuracy of transcribing. Any sections of transcripts that were unclear were checked against the audio recordings. NVivo software [52] was used to code, store, sort and retrieve results from de-identified transcripts. Thematic analysis networks [53] was employed. Following Attride-Stirling [53], ST and CGR independently developed thematic coding manuals using the a priori selected theoretical model (COM-B) as a guide but being open to new codes emerging. In developing the manual, two iterations of coding took place with the two researchers each coding five transcripts to identify themes and relevant statements or quotes. Codes were organised into sub-themes and broader conceptual themes. The coding manual was revised and discussed after each iteration until both researchers were in agreement. These two investigators (ST and CGR) then independently coded all of the interviews. Any discrepancies in the coding manual and codes were resolved through discussion. The researchers used statistical measures of inter-coder verification using the Coding Comparison query in NVivo to identify the reliability of the study. This function calculates the percentage agreement between the two coders, which is the number of units of agreement divided by the total units of measure within the data item, presented as a percentage. Ten interviews were selected to conduct the coding comparison query including five from the interviews conducted with mothers that have not yet introduced solids and five from mothers that have introduce solids to their infant.

Inter-rater reliability ranged from “poor” (Kappa <0.40) for the target behaviour best practice formula feeding (possibly due to the small number of participants who formula fed [54]) to “excellent” (Kappa >0.75) for the target behaviours age of solids introduction and healthy first foods with the remaining target behaviours being rated as “fair to good’ (Kappa 0.40 < 0.75) [52].

A summary of the main findings is contained in Table 3 and a description of the findings for each of the target behaviours is provided below.

Initiation of breastfeeding or formula feeding began with a mother’s motivation (Reflective Motivation) to either breastfeed or formula feed [She was going to be breastfed no matter what. Breastfeeding mother 10]. This desire or plan often formed early – either in pregnancy or even before pregnancy. Some mothers had never considered an alternative to breastfeeding [“I don’t think there was ever a time when I wasn’t going to breast feed,” breastfeeding mother 1]. Reasons for planning to breastfeed were that it was broadly perceived as being nutritionally optimal for the infant [Just because I knew it was good for her and I wanted to do wanted to do what was best for her and I wanted to do what my body is made for. Breastfeeding mother 19], for bonding, health (e.g., immunity), ‘naturalness’, convenience and cost […just that’s what our breasts were made for so you may as well use them, and it’s free as well I guess, less hassle of doing bottles and having to spend extra money when you don’t have it. Breastfeeding mother 2]. However, other mothers were not motivated to breastfeed for reasons such as it feeling unnatural or strange [Very uncomfortable. It’s strange, but yes, I definitely didn’t want to breastfeed at all. It definitely made me very uncomfortable and I didn’t breastfeed either of my children. Formula feeding mother 2]. Finally, other mothers took a pragmatic approach whereby they planned to breastfeed but were aware that it ‘didn’t always work out’ [Like I just wanted to try to be really relaxed, and if it worked it worked, and if it didn't I wasn't - like I was determined not to feel like a failure if I couldn't breastfeed. Breastfeeding mother 3]. Mothers who had previously breastfed a baby were often more motivated to breastfeed (Reflective Motivation) and possessed more skills in breastfeeding (Physical Capability) as well as knowledge about how to breastfeed (Psychological Capability) “I think being my second child breastfeeding, she’s just been very good at latching on and feeding since she was born, which is different to my first experience…the first few months, even though it was a lot easier” Breastfeeding mother 21]. Whereas those who had previously had difficulties breastfeeding a child were the reverse “it was my preference to breastfeed. But because I'd had trouble with my first baby…I was also a bit realistic in that it might not be an option for me. As it turned out, it wasn't an option for me” Breastfeeding mother 1]. Likewise, those who had a positive experience of formula feeding a previous infant (Behaviour) were also more likely to be Motivated to formula feed again [Yeah, I never even considered breastfeeding with my second because I had such a good experience with bottle feeding with my first…so I decided to go the same way again. Formula feeding mother 2].

For those mothers who were motivated to breastfeed, Physical Capability (breastfeeding skills) as well as Psychological Capability (mental toughness, determination) affected whether they took up breastfeeding after the baby was born. For example, this mother struggled with her infant’s reflux and weight loss and was advised that formula would help: [But yes I mean I would have loved to give him all the benefits of the immune system and my health benefits and everything but it just wasn't suitable. Formula feeding mother 3]. For instance for some mothers who had planned and wanted to breastfeed (Reflective Motivation), but experienced problems with latching or mastitis for instance (Physical Capability) this aroused negative emotions (Automatic Motivation) and reduced likelihood of them breastfeeding (Behaviour) [after a month of breastfeeding I did give up after having mastitis three times and also suffering with post natal depression, it just wasn’t something that worked for me. Formula feeding mother 5]. Mothers who were high in Mental Capability and/or were motivated (Reflective Motivation) were able to get through this difficult period and establish breastfeeding [There was a stage where breastfeeding was hard and I was contemplating stopping, but I couldn’t bring myself to do it because I felt like it’s wrong to give him formula, like it’s not natural, like it’s a man-made thing and I want him to be as healthy and to grow up with the best possible start. Formula feeding mother 3].

The main influences on the duration of breastfeeding appeared to be Reflective Motivation, relating to beliefs about the benefits of breastfeeding to the baby and to the mother (convenience and ease) as well as mothers’ plans or goals to achieve a minimum duration of breastfeeding [Your body’s got everything that your baby needs… there are so many different types of bacteria and stuff in breast milk…But we can’t make those in the formula…, I can just go out with my baby and just stop and breastfeed for a second and all these sorts of things, yeah, whereas getting bottles and formula and stuff like that, it does cost a lot of money and it’s good. Breastfeeding mother 7]. Social Opportunity (norms) seemed to have less of an influence on this behaviour. Representing an Environmental Opportunity barrier, returning to work was often the impetus for stopping breastfeeding […I’m going back to work when he’s nine months old so I’ll probably feed him until probably seven months, so I can get him on the bottle before I go back to work. Formula feeding mother 2]. The influence of the infant on the mother was also important. Mothers were physically unable to continue breastfeeding (Physical Capability) when their infant self-weaned [She’s pretty much self-weaning, she’s not really interested. Mixed feeding mother 1], whilst others were motivated (Reflective Motivation) to continue breastfeeding because their infant did not like to drink milk from a bottle [I don't have a choice, they don't like bottles. Breastfeeding mother 11]. Advice (Environmental Opportunity) did not appear to have an influence on breastfeeding duration in that advice on breastfeeding appeared to be given to mothers only during pregnancy or just after birth.

Social Opportunity and Environmental Opportunity were barriers towards best practice formula feeding: Mothers mentioned there was little support and information available to those who formula feed their infant. Furthermore, mothers reported that they felt judged and unsupported by health professionals who were perceived to be pro-breastfeeding [I mean, when you first have a bub you’re thrown into, I suppose, breastfeeding and you’re given so much advice and so much support based on that, but if you have to change to formula or something like that, it’s very negatively viewed upon, even by health practitioners. Formula feeding mother 5]. For this reason advice and information on how to formula feed came primarily from the formula tin, and through online searches. Most mothers appeared to avoid putting their infant to bed with a bottle of formula and routinely followed instructions on tin about formula preparation, adjusting the volume of formula to their infant’s hunger levels [No. I mean I was breastfeeding for the first 6 months, so I used the Australian Breastfeeding line for advice on that. With the formula, I just go off the instructions off the formula bottles and off her cues as well. Formula feeding mother 1]. Mothers were confident in their ability to formula feed their infants (Psychological Capability). Whether others were formula feeding in their social network didn't appear to have much influence.

Reflective Motivation emerged as the main barrier towards introducing solids in line with Australian Infant Feeding Guidelines [55]: Participants rarely mentioned a desire to wait until their infant was six months of age before introducing solid foods [No, I don’t think it’s [waiting until 6 months] realistic at all. Every baby’s different and if we had of waited for her to be six months, she wouldn’t have been very happy at all. Formula feeding mother 2]. This is a rare quote from a mother who was in favour of waiting: “yeah I think so…other mum’s will say I had to start earlier because they were looking at my food and wanting to put it in their mouth and I sort of think, well, babies look at everything and want to put it in your mouth” Breastfeeding mother 5]. In contrast, Reflective Motivation to engage in the competing behaviour (introducing solids early) was higher. The reasons were related to mothers’ beliefs about the consequences of the behaviour (it was perceived as beneficial to the infant to introduce solids early, for example reducing hunger, sleeping longer) [I don't think it's realistic [waiting until 6 months], because if a baby shows that they're ready, I think just go with what your baby's telling you. Because instead of being like them wanting more and more feeds - it's breaking their sleep as well, and they're not getting any sleep. Breastfeeding mother 6]. Mothers mentioned several potential benefits of introducing solid foods earlier, yet appeared to have few beliefs about possible negative impacts to their infants of introducing solid foods earlier, and affected the age at which solid foods were introduced. Mothers were also motivated (Reflective Motivation) to introduce solids when they perceived their baby to be ready (indicated by signs/cues), rather than based on health recommendations [I think it’s [government recommendation] open to interpretation in the fact that okay, each parent knows their child best and every child develops differently. And if some children need to have solids earlier, then who’s to say that they can't? Formula feeding mother 7]. Furthermore, although most mothers appeared knowledgeable about the recommended age at which solids should be introduced some mothers were confused (Psychological Capability) about when to introduce solids [I guess the information on when to start solids is probably more confusing than the breastfeeding information almost. It's like - because it does seem to change a bit but - and then I've heard that if you start them - the earlier you start them the less likely they are to have allergies but then I don't know whether that's true or not. Breastfeeding mother 8]. Social pressure or social norms (Social Opportunity) was also important with mothers recounting receiving pressure or advice from family members or peers to introduce solids at early ages which some mothers chose to ignore […my mother-in-law suggested that I start giving him solids at two months, so I think that’s the older way of going about doing things which I absolutely refused to do. Formula feeding mother 5].

Not surprisingly, mothers were Motivated (Reflective and Automatic motivation) to give those foods that they believed was best for their baby (what their baby needed or wanted) [I look at the way she is. I started her on purees and I could tell that she wasn’t interested in her food anymore, so I tried something different, like mashed food. Now she’s a bit over it, so I’m trying finger food. Breastfeeding mother 9]. Mothers reported having made plans about which foods they would like to introduce and which they would like to avoid, often relying upon heuristics such as ‘fresh foods’, ‘no packaged foods’, ‘no sugar’, ‘homemade’, ‘fruits and vegetables’ [I guess I've always thought fresh is best. So I always try where I can to give him fresh food, wholesome food. Formula feeding mother 6]. Mothers were also motivated to avoid allergenic foods and those that may pose a choking hazard [I've heard that if you start them - the earlier you start them the less likely they are to have allergies but then I don't know whether that's true or not. Breastfeeding mother 4]. Confidence (Psychological Capability) was affected by past Behaviour (having had a child previously) their infant’s reactions (e.g., eating the food, gaining weight) and further experience of feeding their infant (more time after introducing solids). Confidence, as well as knowledge (Psychological Capability) was negatively affected by receiving confusing advice (Opportunity) about which foods to give to infants at different stages of development and with inexperience (early on in the introduction of solids period) [This is where I get confused as well, because people say you need to start on fruit first. Some people say Farex, and other people say vegetables. Breastfeeding mother 3]. For example, mothers relied upon online blogs, popular books, Google searches and Facebook, or on family/friends as a source of information about which foods to give infants in the absence of other reliable and timely information from health professionals [No, only on Facebook group that the mums were talking about what they were going to be introducing to their kids. But other than that, just a hundred percent reliant upon the book really. I kind of take what the Facebook group says with a bit of a grain of salt sometimes Formula feeding mother 4].

Mothers’ desires (Reflective Motivation) to use non-nutritive feeding (primarily feeding to settle) appeared to be higher than desires to feed according to their infant’s appetitive cues. A desire, plan or perceived need to avoid non-nutritive feeding was absent (Reflective Motivation). Mothers often said that they would do ‘whatever works’ to settle the baby and this often included offering milk to their infant. Furthermore, mothers’ beliefs about the consequences of using milk to settle were positive, as it was perceived as an effective settling technique (Reflective Motivation). There was very little indication that using milk to settle the infant would have any negative consequences for the infant (Reflective Motivation). [He will just follow me around, like he crawls, just crying at me until I give him a biscuit or a bottle. Then he's fine, as long as he's like been given something he's happy. Formula feeding mother 2]. Social Opportunity was also a barrier towards mothers’ use of feeding to appetite: Mothers recounted that although at times they were given advice on how to settle their infant without milk from health professionals, family or peers, they were given little support or advice on or information about the possible negative consequences for the infant of using non-nutritive feeding […we were told that we were doing the wrong thing with (baby’s name) by cuddling or feeding her to sleep. Breastfeeding mother 3]. However, other aspects of Motivation, such as making plans about feeding to appetite or to settle were largely absent from the discussion about feeding to appetite/use of non-nutritive feeding.

Aside from using milk/food to settle the infant, many mothers did report allowing their infant to stop feeding when full (Behaviour) [No I just purely go on if she's eating it, like I go on her cues. If she's full, if she's not interested, then that's enough. Sometimes she might not even eat any of it. Formula feeding mother 1]. This was largely affected by Reflective Motivation: mothers believed that the infant was able to determine if she or he has full [He'll pull off the bottle, his head will turn to the side and he just won't latch back on so we just don't - we offer it to him. If he doesn’t want it - he knows his own body more than we do. Formula feeding mother 3], although there were also mothers who tried to get their infants to eat a certain amount of food/milk, believing that the infant required more food than he/she wanted [Facilitator: Yeah, that’s right. So with the formula, is there a particular amount that you do try to give to him every day? Interviewee: Anything over the 500 mark, as long as he’s having roughly 500mls a day. Formula feeding mother 7].

The data supporting the findings are contained within the manuscript and tables.