Assessment of infant feeding styles among low-income African-American mothers: Comparing reported and observed behaviors

Abstract

This study's goal was to provide a detailed description of feeding styles adopted by a sample of African-American women in feeding their infants in North Carolina, and to examine the correspondence between reported and observed feeding styles. Cross-sectional semi-structured interview and videotaped data were gathered in the homes of 20 participating low-income mothers of infants aged 3–20 months. Feeding styles were characterized through a tailored coding scheme (the Infant Feeding Styles Video Coding Scheme, IFSVCS) applied to both interview and video-taped data. We found that the most frequent feeding styles identified for both interviews and videotaped observations was restrictive, but that mothers were roughly equally divided among predominantly controlling (pressuring or restrictive) and less controlling (laissez-faire or indulgent) styles across methods. However, for over 2/3 of the sample, there was a lack of correspondence between interview and video-taped feeding styles. This unique characterization and comparison of observed and reported infant feeding styles provides additional insights into parental feeding approaches among mothers of infants at high risk of obesity, and highlights the need for further study of feeding style assessment and potential impact on infant weight outcomes.

Introduction

Parental feeding styles, or the parental attitudes, beliefs and/or practices related to child feeding, have recently received substantial attention in the literature (Faith, Scanlon, Birch, Francis, & Sherry, 2004; Golan & Crow, 2004; Hughes, Power, Fisher, Mueller, & Nicklas, 2005) for the potential role they may play in the rising epidemic of childhood obesity (Hedley et al., 2004; Strauss & Pollack, 2001). Building on broader parenting style research, Costanzo and Woody (1979) initially suggested that parents may adopt domain-specific parenting styles in relation to different aspects of parenting, such as in relation to discipline, schooling or feeding. They also proposed that certain parenting styles adopted during feeding could be associated with the development of childhood obesity (Costanzo & Woody, 1979). This led the way for other researchers to begin further exploring the association between parenting styles and feeding outcomes (Birch, McPhee, Shoba, Steinberg, & Krehbiel, 1987; Cullen et al., 2001; Gable & Lutz, 2000). Researchers have generally limited their focus to aspects of the authoritarian parenting style during feeding (i.e., parental feeding control) and its association with child feeding behavior and weight status among small samples of predominantly white, middle-class, school-aged children (Birch & Fisher, 2000; Carper, Fisher, & Birch, 2000; Faith, Berkowitz et al., 2004; Faith, Scanlon et al., 2004; Fisher & Birch, 1999; Klesges, Stein, Eck, Isbell, & Klesges, 1991). A recent review of these studies by Faith et al. (Faith, Scanlon et al., 2004) highlighted that it is particularly restrictive feeding strategies (rather than general parental control), which were most often found to be positively associated with increased child energy intake or body weight, although it was not clear if feeding styles play a causal role in the development of childhood obesity, or if those styles emerge as a consequence and then aggravate the problem. The generalizability of the findings to other ethnic groups is also unclear, since 2 larger multi-ethnic studies failed to find similar associations (Baughcum et al., 2001; Robinson, Kiernan, Matheson, & Haydel, 2001). Because obesity has been found to be rising at an alarming rate among preschool-aged African-American and Hispanic children (Hedley et al., 2004; Strauss & Pollack, 2001), the need for further study in this area is evident, especially among younger children in minority populations. Several recent studies have addressed some of the previous studies’ limitations by broadening their scope to include other feeding styles (Hughes et al., 2005; Patrick, Nicklas, Hughes, & Morales, 2005) and focusing on younger (Baughcum et al., 2001; Fisher, Birch, Smiciklas-Wright, & Picciano, 2000; Taveras et al., 2004), and minority, low-income populations (Baughcum et al., 2001; Hughes et al., 2005; Patrick et al., 2005). Additionally, though previous research used different methods to assess parent feeding practices, such as in-home observations and questionnaires (Faith, Scanlon et al., 2004), none, to our knowledge, has yet published any data on the correspondence between these two modes of assessment. One study did show that mothers report different responses related to feeding styles when assessed by questionnaire or semi-structured interview, raising a concern about the validity of questionnaire-type data (Jain, Sherman, Chamberlin, & Whitaker, 2004). Another study indicated that parental report of increased restriction in child feeding was not reflected in their daughters’ reports, generating further questions regarding the occurrence of reported behaviors in actual interactions (Carper et al., 2000).

The present study focuses on a broader range of feeding styles in a sample of low-income African-American infants, drawing on our own previous research and experience in infant feeding (Engle, Bentley, & Pelto, 2000; Ha et al., 2002; Lederman et al., 2004). A comparison of observed and reported feeding styles using qualitative methods provides rich detail on infant feeding styles in this population, within the framework of the parenting style classification proposed by Maccoby and Martin (1983).

The feeding style definitions employed in this study were chosen a priori, and formed the basis of all the analyses conducted in this paper. They are defined as follows: (1) Responsive (in which parents are responsive to the infant's hunger and satiety cues but control the quality of their infant's diet by providing an array of high-quality foods); (2) Pressuring (in which parents are not responsive to their infant's satiety signals and are intent on controlling the amount of food the child gets by increasing the amount consumed); (3) Restrictive (in which parents are not responsive to their infant's hunger signals and are intent on controlling the amount and quality of food the child gets by decreasing the amount consumed and/or not allowing the child to eat lower-quality foods); (4) Indulgent (in which parents are responsive to hunger and satiety cues but do not control or set limits in terms of the quantity or quality of food consumed); and (5) Laissez-Faire (in which parents are not responsive to hunger and satiety cues and do not control or set limits in terms of the quantity or quality of food consumed). These styles can also be thought of in terms of the general levels of parental control exerted during the feeding, ranging from controlling (pressuring or restrictive) to less controlling (indulgent or laissez-faire).

The aims of the present study were: (1) to verify the presence of the 5 theoretical feeding styles, as defined above, in a population of low-income African-American mothers of infants under the age of two, using a specifically tailored coding scheme for both semi-structured interviews and video-taped observations; and (2) to examine the correspondence between reported (semi-structured interview) and observed (video-taped) predominant parent feeding style data collected in the families’ homes.