Feeding Sensitivity at 3-Months Predicts Parental Feeding Style at 3-Years

Objective: To examine the association between maternal responsive feeding in infancy and later parental feeding style in childhood, with a secondary aim of linking parenting style to child weight status Design: Short-term, longitudinal study Subjects: 65 mother-infant pairs, followed from 3-months to 3-years Methods: Mother-child dyads were seen at enrollment, observed at home when the infant was 3-months-old, and at home again when the infant turned 2-years of age. When the child reached 3-years, mothers self-reported their feeding style. Results: Responsive maternal feeding at 3-months predicted an authoritative feeding style at 3-years, and the authoritative feeding style was associated with a healthier weight status. Conclusion: Responsive feeding in infancy may serve as the foundation for responsive feeding in childhood and may serve to help in reducing obesity risk

To this point, a number of instruments are now available that allow for caregiver self-reports as to how she approaches feeding her young child [5][6][7][8][9]. While early efforts largely focused on caregiver control as expressed through restricting children's intake or pressuring them to eat (see [10] for a review), more recent reports have used the feeding context as a window to parenting style in general (see [11] for a review). For example, parental restricting and pressuring has been shown to predict the authoritarian parenting style [12]. The authoritarian style-indeed, the most controlling-reflects high demandingness on the part of the parent, but also low responsiveness to the child's needs [13]. Three other parenting styles have been formulated, namely, authoritative (high demandingness and high responsiveness), indulgent (low demandingness and high responsiveness), and uninvolved (low demandingness and low responsiveness) [14]. The authoritative parenting style is acknowledged to be the best in terms of the child's developmental outcomes [15], and current research seems to support its being the most protective style in reducing obesity risk [11].
As sizable as the research in this area has become, however, it is remiss in ignoring the role of responsive parenting during the child first years, that is, how the caregiver handles feeding during infancy. Though not considered a form of parenting style, responsive feeding is nonetheless an index of responsive parenting [16], with nonresponsive feeding having demonstrated an association with child overweight [17]. Fortunately, instruments that can measure feeding behavior during infancy are also available, both as selfreport [5,18] and notably, via observation (see [19] for a review).
The purpose of the present study was to explore the association between maternal responsive feeding in infancy and later parental feeding style in childhood, with a secondary aim of linking parenting style to child weight status. To this end, maternal behavior during a feeding interaction was observed at 3-months, and mothers later self-reported their feeding behaviors at 3years, with the latter used to categorize them along the lines of recognized parental feeding styles.

METHOD
The research protocol was approved by the Rutgers University Institutional Review Board. As part of the Rutgers Infant Nutrition and Growth (RING) Project, mothers were recruited at their initial visit to the local WIC Center. WIC is the Special Supplemental Nutrition Program for Women, Infants, and Children [20] that provides funds to states for supplemental foods and nutrition education for low-income pregnant and postpartum women, and to their infants and children up to age five who are deemed to be at nutritional risk. The cooperating center serves a largely minority population. The RING Project was designed to explore risk factors for early obesity, hence only mothers who indicated they were formula-feeding their infants were approached to participate in the study. Following their enrollment, the mothers were queried as to their race/ethnicity and provided some demographic information and details concerning their infants' birth. All the participants were non-white, specifically Black, Hispanic, or other Latino, with Spanish translation provided by the recruiter where necessary. The present sample consists of 65 mother-child pairs who were seen at enrollment, at home when the infant was 3months-old, and at home again when the infant turned 2-years of age. When the child reached 3-years, mothers completed a questionnaire.

Measures
When the infants reached 3-months of age, two research assistants (one Spanish speaking as needed) visited the family's home. Infants were measured and weighed twice using a pediatric scale (Tanita TM BD-585) and mother-infant pairs were observed using the Nursing Child Assessment Feeding Scale (hereafter, CAFS)-a 76-item rating scale of mother and infant interactive behaviors coded during a feeding episode.
The CAFS is well-validated and suitable for infants from birth-12 months [21]. When the infants reached 2years of age, two research assistants returned to the families' homes, with the infants again being measured and weighed with the pediatric scale. When the child was 3-years-old, mothers were sent the Caregiver's Feeding Style Questionnaire (CFSQ) to complete and return. The CFSQ is a 31-item instrument that taps the physical and verbal strategies a mother uses to get her child to eat, and reflects the dominant parental feeding style. The CFSQ was developed for and validated with parents of 3-5-year-old children [8].

CAFS
For the CAFS, subscale scores for the mother are provided for her Sensitivity to Cues, Response to Children's Distress, Social-Emotional Growth Fostering, and Cognitive Growth Fostering, and a Caregiver Total is then computed. A higher Caregiver Total Score therefore denotes a mother who is higher in sensitivity and responsiveness to her infant in the context of feeding.

CFSQ
For the CFSQ, subscales are computed for the dimensions of Demandingness (D) and Responsiveness (R), and using median scores for higher and lower, the caregiver can be classified as exhibiting an authoritative (high D, high R), authoritarian (high D, low R), indulgent (low D, high R), or uninvolved (low D, low R) parental feeding style. Table 1 provides descriptive statistics for the mothers and infants as obtained at the time of their recruitment. The average mother had less than a high school education. Their self-reported weight and derived Body Mass Index (BMI) indicates that over half were overweight prior to becoming pregnant. Approximately half of the mothers were of Mexican origin, with the remainder split between Black and Latina. Table 2 displays the CAFS scores derived from the observations made at the 3-month home visit, as well as the weight measures collected at 3-months and 2years.

RESULTS
As shown in Figure 1, the highest Caregiver Total at 3-months was associated with the authoritative feeding style at 3-years, with a score of 38.33. In contrast, the other three parental feeding styles hovered around a total score of 33, as the authoritarian, indulgent, and uninvolved styles were at 32.25, 34.25, and 33.00 respectively. As weight and height measures were obtained for the children at 2-years, BMI for age and sex were calculated for each child [22].
As shown in Figure 2, children whose mothers were categorized as authoritative with the CFSQ had the most optimal BMI for age and sex, with their average being closest to the 50th percentile.

DISCUSSION
In the present study, mothers were observed interacting with their 3-month-old infants during a feeding bout, and were scored with respect to their sensitivity and responsiveness. When their infants reached 3 years of age, the same mothers completed a questionnaire that allowed them to be classified as to their feeding style, along with the lines of how demanding and responsive they were while feeding their child.
While the literature on parenting styles has a relatively long history, with the authoritative style acknowledged as the most optimal, linkages to  outcomes such as child BMI have only recently been explored [11]. The present result, that an authoritative feeding style was associated with the healthier weight status, that is, child BMI at the 50th percentile, is in harmony with studies suggesting that authoritative parenting may lower the risk of children becoming overweight or obese [23][24][25]. Conversely, the finding that the uninvolved or neglectful style was associated with 2-year-olds who were at the 80th percentile for their BMI is supported by both longitudinal [26] and cross-sectional work [27].
Beyond the associations between parental feeding styles and child BMI, how infant feeding is approached by the caregiver before a feeding style per se emerges is also important. In the present study, responsive feeding at 3-months was the bellwether of an authoritative feeding style at 3-years. As the authoritative style is largely defined by high responsiveness, these results suggest that responsive feeding in infancy may serve as a foundation for responsive feeding in childhood, and if coupled with reasonable demands of the child, may reflect optimal parenting in other contexts.
Despite these straightforward findings, limitations of the present investigation must be acknowledged. The sample was comprised of low-income, minority mothers, who chose to exclusively formula-feed their infants, rendering as only tentative their comparability to most studies on parenting and feeding style that have been largely conducted on white, middle-class samples. The WIC mothers in the current sample were also bereft of secondary education. However, available  research on low-income, minority families have shown feeding styles such as indulgent and uninvolved to predominate [8,28], suggesting that the present results are valid.
Of greater concern might be that the maternal selfreports of parenting style were obtained a year after the anthropometric measures were made, raising the issue that the child's weight status may have influenced how she rated herself. However, evidence exists that parenting styles are relatively stable over the first 6 years of the child's life, suggesting that parental responsiveness, in particular, would not differ fundamentally from 24-to 36-months, and would likely have been similar if rated as early as 15-months [29].
In sum, the present study suggests that a sensitive and responsive maternal feeding style, evident in the first months of the infant's life, may be an indication of how responsive the mother will be as her child grows beyond toddlerhood. Although this investigation obtained data at different time points, longitudinal studies that observe feeding style repeatedly are certainly warranted. But given the associations were shown between feeding style and the child's weight parameters, interventions that seek to reduce the risk for childhood obesity should incorporate education for responsive feeding as a matter of course.