- Email: [email protected]
Parent–child interaction during feeding or joint eating in parents of different weights
Eating Behaviors 18 (2015) 131–136
Contents lists available at ScienceDirect
Eating Behaviors
Parent–child interaction during feeding or joint eating in parents of different weights Verena Wendt a,b, Sarah Bergmann a,b, Katharina Herfurth-Majstorovic a,b, Anja Keitel-Korndörfer b, Kai von Klitzing b, Annette M. Klein b,⁎ a b
Leipzig University Medical Center, IFB AdiposityDiseases, Leipzig, Germany Department of Child and Adolescent Psychiatry, Psychotherapy and Psychosomatics, University of Leipzig, Leipzig, Germany
a r t i c l e
i n f o
Article history: Received 11 December 2014 Received in revised form 2 March 2015 Accepted 29 April 2015 Available online 10 May 2015 Keywords: Parent–child interaction Child feeding Mealtime interaction Obese mother Obese father
a b s t r a c t The current study investigates parent–child interaction during feeding or during joint eating, and aimed to explore differences in feeding interactions between mothers and fathers, as well as between overweight, obese and not overweight parents. 148 mothers and 148 fathers with children aged between 7 and 47 months were observed during feeding of or joint eating with their child in the laboratory. The videotaped mother–child and father–child dyads were coded using the Chatoor Feeding Scale. This scale consists of 5 subscales: Dyadic Reciprocity, Dyadic Conflict, Talk and Distraction during Feeding, Struggle for Control, and Non-Contingency. Compared to mothers, fathers showed higher readings on the Talk and Distraction scale; in all other subscales no differences were found. The comparison between overweight, obese and not overweight mother–child dyads revealed no significant differences. Differences in father–child dyads between overweight, obese and not overweight fathers were identified in the subscale Struggle for Control: overweight fathers were marked by a higher amount of Struggle for Control than obese and not overweight fathers. Taken together, differences found in the present observational study are small to moderate, and thus the current results support extant literature demonstrating that there are no differences in feeding behaviour between mothers and fathers or between obese and non-obese parents. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction The development of children's weight is influenced by many factors. While research investigating factors that contribute to overweight or obesity has previously focused on the individual, in recent decades familial factors and parent–child-interactions have received more attention. Research on feeding is unanimous that there is a connection between certain feeding practices (e.g. restriction) or maternal decisions of what to feed (e.g. to breastfeed the child), and overweight or obesity in childhood (e.g. Arenz, Rückerl, Koletzko, & von Kries, 2004; Bergmeier, Skouteris, Horwood, Hooley, & Richardson, 2014; DiSantis, Hodges, Johnson, & Fisher, 2011; Faith & Kerns, 2005; Faith, Scanlon, Birch, Francis, & Sherry, 2004), but critical gaps remain in knowledge about differences in feeding behaviour between mothers and fathers in general, and differences between obese and non-obese parents. Although the time fathers spend interacting with their children has in general increased (Yeung, Sandberg, Davis-Kean, & Hofferth, 2001), research on the feeding practices of fathers is still scant. A review by ⁎ Corresponding author at: Department of Child and Adolescent Psychiatry, Psychotherapy and Psychosomatics, University of Leipzig, Liebigstrasse 20a, 04103 Leipzig, Germany. Tel.: +49 341 9724018; fax: +49 341 9724019. E-mail address: [email protected] (A.M. Klein).
http://dx.doi.org/10.1016/j.eatbeh.2015.04.007 1471-0153/© 2015 Elsevier Ltd. All rights reserved.
Khandpur, Blaine, Orlet Fisher, & Davison (2014) considered studies that have included fathers. They found 20 studies assessing fathers' child-feeding practices via self-reports (18 studies) or via mealtime observations (2 studies). It was shown that, compared to mothers, fathers control and limit children's food intake and food access less. Instead, pressuring the child to eat was reported as a common feeding style in fathers (Khandpur et al., 2014; Orell-Valente et al., 2007; Pulley, Galloway, Webb, & Payne, 2014). In fathers, concern that the child could become overweight is associated with higher responsibility for child feeding and monitoring of what and how much the child eats (Mallan et al., 2014). However, research findings comparing the feeding pattern of parents are not consistent. A number of studies have found no differences between mothers and fathers in terms of restrictive or pressuring feeding patterns (Blissett, Meyer, & Haycraft, 2006; Haycraft & Blissett, 2008). Although there is a strong association between maternal obesity and heightened risk of obesity in childhood, feeding differences between obese, overweight and normal-weight mothers have received comparatively little attention in the literature. Thompson (2013) gives an overview of the feeding behaviour of obese mothers. Compared to normal-weight women, overweight and obese women are less likely to breastfeed their children, and more likely to wean them earlier (Amir & Donath, 2007). Haycraft and Blissett (2008) reported that
132
V. Wendt et al. / Eating Behaviors 18 (2015) 131–136
obese mothers show more restriction and control of the quantity and quality of foods provided. Furthermore, more pressure to eat during mealtime was observed in women with a higher body mass index (BMI) (Haycraft & Blissett, 2008). However, comparisons of feeding styles between obese and non-obese mothers are partially contradictory: no differences in mother–child feeding behaviour (Lewis & Worobey, 2011; Lumeng & Burke, 2006; Lumeng et al., 2012), especially after controlling for family income (Baughcum et al., 2001), or less control by obese mothers (Wardle, Sanderson, Guthrie, Rapoport, & Plomin, 2002), have also been shown. The current study therefore investigates parent–child interactions during feeding or joint eating. The central aims of the study are to examine differences (1) between mothers and fathers, and (2) between obese, overweight and not overweight mothers and fathers. As there is either a lack of research or the results in literature are inconsistent, we investigated possible differences between mother–child and father–child feeding interactions as well as between overweight, obese and not overweight mothers and fathers, in an exploratory fashion. 2. Method 2.1. Study design and sample The current study is part of the Integrated Research and Treatment Center (IFB) AdiposityDiseases at the University Medical Centre Leipzig, and was embedded in a prospective longitudinal study investigating children's psychological and physical development (Grube et al., 2013). Families were recruited via information and flyers left at kindergartens, paediatricians' practices and facilities for children and families in the area of Leipzig, Germany. Out of a larger sample, this study included families in which both parents are involved in child care and took part in a feeding/joint eating observation with their child in the laboratory. Fathers did not have to be biologically related to the participating child but could be social fathers/ stepfathers. The sample consists of N = 148 mothers and N = 148 fathers with their children (76 girls, 51.4%) aged 7 to 47 months. The participating parents were generally highly educated; for details and further information see Table 1. Out of 148 mothers and 148 fathers, n = 44 mothers and n = 37 fathers were obese (BMI ≥ 30), n = 17 mothers and n = 32 fathers were overweight (BMI ≥ 25.00 ≤ 29.99), n = 83 mothers and n = 77 fathers were normal weight (BMI ≥ 18.50 ≤ 24.99), and n = 4 mothers and n = 2 fathers were underweight (BMI b 18.50). In this study, we combined the groups of underweight and normal-weight participants, named “not overweight” (n = 87 mothers and n = 79 fathers). For the mean BMI in the different groups see Table 1. 2.2. Procedure The families were informed about the study as well as the procedure, participated voluntarily, and each parent gave informed consent prior to their inclusion in the study. The study was approved by the ethics committee of the University of Leipzig. Parents completed a set of questionnaires at home. They were invited to the laboratory of the Department of Child and Adolescent Psychiatry, Psychotherapy and Psychosomatics, University of Leipzig. The appointment was scheduled to coincide with the child's usual mealtime. Mothers or fathers were advised to bring meals the child normally ate at that time. If mothers and fathers also eat while they feed their child at home, they were encouraged to bring food for themselves and to eat with their child in the laboratory. After a warm-up interview, the father or the mother was left alone in the room with the child. They were instructed to feed their child or eat with their child as they normally would at home. Parents were free to position their child as they wished, on their lap, in a highchair
Table 1 Sociodemographics of the t1 baseline sample. Parent characteristics
Mothers N = 148
Fathers N = 148
Age in years, M (SD) Age range BMI, M (SD) Weight group, BMIM(SD) Not overweight Overweight Obese
31.44 (4.27) 21.44–43.12 26.86 (7.52)
34.76 (6.03) 22.52–56.59 26.72 (6.12)
21.87 (1.94) 26.43 (1.34) 36.88 (5.69)
22.44 (1.82) 27.33 (1.37) 35.34 (5.24)
Educational level No degree Certificate of general or secondary educationa General qualification for university entranceb University degree
0 (0.0%) 52 (35.1%) 38 (25.7%) 58 (39.2%)
3 (2.0%) 55 (37.2%) 39 (26.4%) 51 (34.5%)
Employed at t1 Yes No Parental leave or other
89 (60.1%) 29 (19.6%) 30 (20.3%)
112 (75.7%) 29 (19.6%) 7 (4.7%)
Relationship statusc Unmarried Married (living together) Married (living apart) Divorced Other
78 (52.7%) 63 (42.6%) 2 (1.4%) 3 (2.0%) 2 (1.4%)
74 (50.0%) 63 (42.6%) 1 (0.7%) 8 (5.4%) 2 (1.4%)
Child characteristics Age in months, M (SD) Age range, months Sex Girls Boys
24.39 (11.24) 7–47 76 (51.4%) 72 (48.6%)
a The certificate of general education is an elementary school diploma, which is obtained on successful graduation from grade 9; the certificate of secondary education is obtained on successful graduation from grade 10. b This group also includes parents with the entrance qualification for a university of applied sciences. c Mothers and fathers are parenting the same child. Differences in the relationship status indicated by mothers and fathers result from e.g. fathers being divorced from an earlier marriage.
or on a normal chair at the table. The feeding interactions were videotaped for about 20 min. Each parent fed the child once. Mothers and fathers determined – according to circumstances – who fed the child first. As the child could not be fed twice successively, a further appointment was arranged for the other parent. 2.3. Measures 2.3.1. Anthropometric data The body mass index (BMI) was calculated using the weight and height of mothers and fathers, measured in the lab in a standardized way. Parents were weighed without shoes on calibrated scales. A calibrated stadiometer was used to measure height. A parental BMI below 18.50 indicates underweight, a value of 18.50 to 24.99 normal weight, of 25.00 to 29.99 overweight, and a BMI above 30 suggests that the person is obese (World Health Organization, 2000). 2.3.2. Mealtime observations The videotaped mother–child and father–child interactions during feeding and joint eating were rated using the Chatoor Feeding Scale (Chatoor et al., 1997), which was developed to assess mother–child interaction during feeding. It was designed for children aged 1 month to 3 years and consists of 26 mother-specific items and 20 child-specific items. These items are arranged under 5 subscales: Dyadic Reciprocity (quality of relatedness and affective engagement between the mother and her child; 11 maternal items and 5 child items), Dyadic Conflict (overt conflicts between mothers and their children over eating;
V. Wendt et al. / Eating Behaviors 18 (2015) 131–136
4 maternal and 8 child items), Talk and Distraction during Feeding (attempts by the dyad to engage or control each other by talking or distracting each other during feeding; 2 maternal and 2 child items), Struggle for Control (efforts or behaviour by mother and child to control feeding; 4 maternal and 3 child items), and Maternal Non-Contingency (the mother's inability to interpret child cues and respond contingently; 5 maternal and 2 child items). All items are rated on a four-point Likert scale (0–3) ranging from “none”, “a little”, “pretty much” to “very much”, and depict frequency or quality of behaviour.
2.3.3. Coding The videotaped parent–child feeding interactions were rated by two trained coders. For assessment of inter-rater reliability, 23% of the videos were coded twice. The inter-rater reliability was highest for Dyadic Reciprocity (mean rICC = .89, p b .001) and for Talk and Distraction during Feeding (mean rICC = .83, p b .001). The inter-rater reliability for the Dyadic Conflict subscale was rICC = .79 (p b .001), for the Struggle for Control subscale rICC = .77 (p b .001) and for the Maternal NonContingency subscale rICC = .71 (p b .001).
2.4. Statistical analysis Data handling and all statistical analyses were carried out using IBM SPSS Statistics (release 20). First, we conducted descriptive analyses (including Chi2 tests) in order to illustrate the feeding/joint eating interactions. Second, we calculated Pearson correlation coefficients to assess associations of the Chatoor subscales. Third, we conducted a repeated measures multivariate analysis of variance (MANOVA), with parent (mother vs. father) as the within-subjects factor and the Chatoor subscales as dependent variables followed by univariate analyses (ANOVAs). Fourth, for assessing differences between obese vs. overweight vs. not overweight mothers/ fathers, we conducted MANOVAs or MANCOVAs with the betweensubjects factor weight group, controlling for sociodemographic variables, if there were at least marginal significant differences (p b .10) between weight groups.
3. Results 3.1. Description of feeding/joint eating interactions On average, feeding or joint eating lasted 16 min (Mmothers = 15.51, SDmothers = 7.40; Mfathers = 16.28, SDfathers = 7.32). Mothers and fathers did not differ in the kind of meal they offered their children (χ2 (4, N = 296) = 7.07, p = .132). 16% of dyads ate breakfast, 16% came for lunch, 15% of mothers and fathers offered mush and 53% of parents provided a snack. In 78% of cases mothers or fathers ate together with their child (n = 115 mothers, n = 117 fathers), whereas in 22% of cases only the child ate or was fed by its mother/father during the videotaped feeding/ joint eating interaction.
133
3.2. Correlations of Chatoor subscales Pearson correlation coefficients for all Chatoor subscales were calculated separately for the mother–child and father–child dyads; see Table 2. The pattern of correlations was quite similar for mother– child and father–child dyads: the highest positive correlations were found between Dyadic Conflict, Struggle for Control, and NonContingency. In contrast, these subscales were negatively related to Dyadic Reciprocity. Moreover, there was a significant positive correlation between Dyadic Reciprocity and Talk/Distraction. 3.3. Comparison of mother–child and father–child dyads in feeding/joint eating interactions In our sample, we found a fairly high mean subscale score for Dyadic Reciprocity, whereas the mean scores for Dyadic Conflict, Struggle for Control, and Non-Contingency were low. The mean subscale score for Talk/Distraction was approximately in the middle of the possible range; see Table 3. The MANOVA with parent as within-subjects factor and the Chatoor subscales as dependent variables revealed a main effect of parent, F(5, 143) = 3.01, p = .013, η2p = .095. Univariate analyses showed that mothers and fathers differed significantly with regard to Talk/Distraction, with fathers showing more Talk/ Distraction during feeding/joint eating. Moreover, there was a trend towards significance (p = .079) in Dyadic Conflict, with fathers again showing higher scores. On all other subscales there were no differences between mothers and fathers (for M, F, p and η2p see Table 3). 3.4. Comparison of mother–child interactions during feeding/joint eating of obese, overweight and not overweight mothers Before analysing group differences, we tested whether obese, overweight and not overweight parents differed in sociodemographic variables. Obese, overweight and not overweight mothers did not differ in terms of their child's sex and age or their own age (p N .429), but did differ in their level of education, F(2, 145) = 3.21, p = .043, η2p = .042, which we therefore included as control variable. The MANCOVA with weight group (obese vs. overweight vs. not overweight) as between-subjects factor, the Chatoor subscales as dependent variables and the level of education as a covariate revealed no significant effect of weight group, F(10, 282) = 0.75, p = .681, η2p = .026. There was no significant effect for the covariate level of education either, F(5, 140) = 0.65, p = .662, η2p = .023. The univariate analyses also showed that obese, overweight and not overweight mothers did not differ significantly in any of the subscales, all p N .123 (for M, SD, F and η2p see Table 4). 3.5. Comparison of father–child interactions during feeding/joint eating of obese, overweight and not overweight fathers Obese, overweight and not overweight fathers did not differ significantly regarding their child's sex, their own age or their level of
Table 2 Pearson correlation coefficients for the Chatoor subscales of the mother–child/father–child dyads (N = 148).
1 Dyadic Reciprocity 2 Dyadic Conflict 3 Talk/Distraction 4 Struggle for Control 5 Non-Contingency Two-tailed. ⁎ p b .05. ⁎⁎ p b .01. ⁎⁎⁎ p b .001.
1
2
3
4
5
1/1
−.40⁎⁎⁎/−.32⁎⁎⁎ 1/1
.20⁎/.32⁎⁎⁎ .07/.05 1/1
−.32⁎⁎⁎/−.31⁎⁎⁎ .58⁎⁎⁎/.62⁎⁎⁎
−.20⁎/−.25⁎⁎ .59⁎⁎⁎/.49⁎⁎⁎
−.03/−.03 1/1
−.06/−.15 .56⁎⁎⁎/.55⁎⁎⁎ 1/1
134
V. Wendt et al. / Eating Behaviors 18 (2015) 131–136
Table 3 Chatoor subscale means (SD) for mothers and fathers (N = 148).
Dyadic Reciprocity (16 items) Dyadic Conflict (12 items) Talk/Distraction (4 items) Struggle for Control (7 items) Non-Contingency (7 items)
Mothers
Fathers
Univariate tests
Range of subscale score (manual)a
M (SD)
M (SD)
F
η2p
0–48 0–36 0–12 0–21 0–21
32.97 (4.10) 2.74 (3.12) 4.59 (1.89) 1.47 (1.58) 0.76 (1.13)
32.93 (4.51) 3.33 (3.58) 5.33 (2.38) 1.47 (1.61) 0.87 (1.31)
0.01 3.14+ 12.47⁎⁎ 0.00 0.67
.000 .021 .078 .000 .005
MANOVA univariate tests, two-tailed; df = 1, 147. a According to the conceptualization given in the manual, the five Chatoor subscales differ in terms of the number of items. Hence, the range of possible subscale scores differs as well. + p b .10. ⁎ p b .05. ⁎⁎ p b .01. ⁎⁎⁎ p b .01.
education (p N .154), but there was a tendency towards significance for their child's age, F(2, 145) = 2.77, p = .066, η2p = .037, which we therefore included as control variable. The MANCOVA with father weight group (obese vs. overweight vs. not overweight) as between-subjects factor, the Chatoor subscales as dependent variables and the child's age as a covariate revealed no significant effect of weight group, F(10, 282) = 1.41, p = .174, η2p = .048, whereas the age of the child was highly significant, F(5, 140) = 17.44, p b .001, η2p = .384. The univariate analyses revealed a significant effect of father weight group for the subscale Struggle for Control, F(2, 144) = 6.19, p = .003, η2p = .079. A post-hoc test (Games-Howell) revealed that father–child interactions during feeding/joint eating by overweight fathers were marked by a significantly higher amount of Struggle for Control than those of both obese and not overweight fathers, whereas those of obese and not overweight fathers did not differ from each other. For all other subscales, we found no significant differences, all p N .336 (for M, SD, F and η2p see Table 5). 4. Discussion The comparison of mothers and fathers revealed no significant differences in the type of meal they offered their child, or in the number of parents eating as well vs. the number who just fed their child. The videotaped mother–child and father–child interactions during feeding and joint eating rated with the Chatoor Feeding Scale revealed differences between both genders in Talk and Distraction during feeding/ joint eating. This behaviour was shown more in father–child dyads than mother–child dyads, with moderate effect size. That means fathers allowed more distraction during feeding while they were trying to guide feeding or joint eating by telling the child what to do or not to do and to eat. Those are specific terms of verbal prompts. However, we found no differences between mother–child and father–child dyads in most of the subscales. Father–child dyads showed equally high Dyadic Reciprocity and equally low Struggle for Control and NonContingency. This is in line with research that indicated no differences in feeding styles between mothers and fathers (Blissett et al., 2006; Haycraft & Blissett, 2008), and it may suggest that the interaction
pattern of fathers during feeding or joint eating is comparable to the behaviour of mothers. Fathers seem as able as mothers to show affective engagement and involvement during interaction, to respect the child's cues, to interpret child cues correctly, and to respond contingently. These results fit with evidence suggesting that mothers and fathers are similar in their interaction styles and that paternal patterns closely resemble maternal patterns throughout development (Lewis & Lamb, 2003). To summarize, while fathers show more verbal prompts (Talk and Distraction), they show the same degree of control on a behavioural level during feeding or joint eating as do mothers. Our study did not confirm findings that fathers are more likely to press their child to eat than mothers (Khandpur et al., 2014; Orell-Valente et al., 2007; Pulley et al., 2014). Comparisons between obese, overweight and not overweight parents showed no significant differences in maternal feeding behaviour, and showed significant difference in paternal feeding behaviour only for Struggle for Control. Overweight fathers scored higher on Struggle for Control than either obese or not overweight fathers. It is possible that overweight fathers are more concerned about their child's weight and therefore try to control feeding. This is in line with the finding by Mallan et al. (2014) that a higher level of concern in fathers is associated with greater monitoring of the child's eating. However, as the multivariate effect of weight group was not significant, this result should be interpreted with caution. Apart from this we did not find any significant differences. Therefore, we concluded that overweight and obese parents show a quality of relatedness and affective engagement during feeding/ joint eating that is comparable to that of normal weight parents. We did not observe either higher levels of Dyadic Conflict or of control in obese compared to not overweight parents. Attempts at distraction were also detected equally in parent–child dyads of overweight and obese parents compared to not overweight parents. Overweight, obese and not overweight parents also show the same level of ability to interpret child cues and to respond contingently. This is in line with data on parent feeding styles, where obese mothers are not reported to be more likely to offer food to cope with emotional distress than normal-weight mothers, or to use food as a reward, or to prompt a child to eat more than he or she wanted (Lewis & Worobey, 2011; Wardle et al., 2002). Based on our results, the assumption of intergenerational transmission
Table 4 Chatoor subscale means (SD) for obese (n = 44), overweight (n = 17) and not overweight (n = 87) mothers.
Dyadic Reciprocity (16 items) Dyadic Conflict (12 items) Talk/Distraction (4 items) Struggle for Control (7 items) Non-Contingency (7 items)
Obese mothers
Overweight mothers
Not overweight mothers
Univariate tests
Range of subscale score (manual)
M (SD)
M (SD)
M (SD)
F
η2p
0–48 0–36 0–12 0–21 0–21
32.95 (4.14) 3.39 (3.38) 4.39 (2.06) 1.77 (1.76) 1.02 (1.32)
33.71 (2.44) 1.94 (2.33) 5.06 (1.34) 1.06 (1.14) 0.41 (0.71)
32.83 (4.36) 2.56 (3.09) 4.61 (1.89) 1.40 (1.54) 0.70 (1.08)
0.33 1.31 0.96 1.48 2.13
.005 .018 .013 .020 .029
MANOVA univariate tests, two-tailed; df = 2, 144.
V. Wendt et al. / Eating Behaviors 18 (2015) 131–136
135
Table 5 Chatoor subscale means (SD) for obese (n = 37), overweight (n = 32) and not overweight (n = 79) fathers.
Dyadic Reciprocity (16 items) Dyadic Conflict (12 items) Talk/Distraction (4 items) Struggle For Control (7 items) Non-Contingency (7 items)
Obese fathers
Overweight fathers
Not overweight fathers
Univariate tests
Range of subscale score (manual)
M (SD)
M (SD)
M (SD)
F
η2p
0–48 0–36 0–12 0–21 0–21
33.43 (3.63) 2.92 (3.37) 5.73 (2.28) 1.16 (1.42) a 0.70 (1.02)
31.75 (3.93) 4.25 (3.83) 5.34 (2.36) 2.38 (2.11) b 1.22 (1.60)
33.18 (5.03) 3.15 (3.56) 5.14 (2.44) 1.25 (1.32) a 0.81 (1.28)
1.10 0.90 0.30 6.19⁎⁎ 0.96
.015 .012 .004 .079 .013
MANOVA univariate tests, two-tailed; df = 2, 144. ab Different superscripts indicate group differences found in post-hoc analyses (Games-Howell). ⁎ p b .05. ⁎⁎ p b .01. ⁎⁎⁎ p b .01.
of obesity risk via the way that obese parents feed their children has to be questioned. Wardle et al. (2002) summarize this: “The stereotype of the obese mother who uses food in non-nutritive ways, so that her child too becomes obese, is more likely to be myth than fact” (p. 459). Most studies comparing feeding patterns between obese and non-obese mothers have used questionnaires. Our results therefore complement previous observational studies of this topic (de Campora, Giromini, Larciprete, Li Volsi, & Zavattini, 2014; Lewis & Worobey, 2011; Rising & Lifshitz, 2005). There are some limitations to this study. The sample may not be representative, as families took part voluntarily. Therefore, we cannot rule out a self-selection bias. The two coders rating the parent–child feeding interactions could not be blinded with regard to the weight status of mothers and fathers. We cannot exclude that this has affected their ratings — however, as we found almost no group differences, we do not consider it very likely. Another limitation is the comparatively wide age range of the children, which was chosen in order to allow recruitment of a relatively large sample size. A further limitation refers to the method used. Although observational data have the advantage of being more objective than self-rated questionnaires, the assessment in the laboratory may have altered the families' behaviour to some degree and this may have compromised the results (Bakeman & Gnisci, 2006). In addition, we analysed parents' behaviour in only one feeding interaction and did not assess how often parents feed during the day. There is also a limitation due to the use of the Chatoor Feeding Scale (Chatoor et al., 1997), which was developed exclusively for mother–child interactions and children from 1 month to 3 years of age. In this study the scale was used for evaluating father–child interactions and with children in a broader age range (max. 47 months) as well. As there are to date no instruments that have been validated for father–child feeding interactions we chose the Chatoor Feeding Scale, which in the main suited the age group of the children. On the other hand, there are several strengths to this study. First of all, we observed parent–child interaction during mealtimes, which is a more objective way of assessing feeding behaviour than using selfreports. In addition, there are almost no studies comparing the feeding interactions of obese, overweight and not overweight parents as well as the feeding interactions of mothers and fathers. This study therefore adds important information.
5. Conclusion In our study, mothers and fathers differed in interaction during feeding or joint eating with their child only in terms of behaviour, including distraction and verbal prompts. Moreover, there were almost no differences between obese and not overweight parents. In general, all parents showed interactions with their child that were characterized by affective engagement and involvement, respect for the child's cues, correct interpretation, and contingent responses to child cues.
Role of funding sources This research is part of the Integrated Research and Treatment Centre “IFB AdiposityDiseases” and is supported by the Federal Ministry of Education and Research, Germany (BMBF, FKZ: 01EO1001). The BMBF had no role in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the decision to submit the paper for publication.
Contributors All authors made an essential contribution to the manuscript. Kai von Klitzing is the principal investigator. He designed the study with Annette M. Klein. Verena Wendt, Sarah Bergmann, Katharina Herfurth-Majstorovic and Anja Keitel were involved in coordination, recruitment and establishment and assessment of the cohort. Annette M. Klein conducted the statistical analysis. Verena Wendt and Annette M. Klein wrote the first draft of the manuscript and all authors contributed to and have approved the final manuscript.
Conflict of interest All authors declare that they have no conflicts of interest.
Acknowledgments Our thanks go to all participating parents and children, as well as to the members of our research team. This study is part of the Integrated Research and Treatment Centre “IFB AdiposityDiseases” and is supported by the Federal Ministry of Education and Research, Germany (BMBF, FKZ: 01EO1001).
References Amir, L.H., & Donath, S. (2007). A systematic review of maternal obesity and breastfeeding intention, initiation and duration. BMC Pregnancy and Childbirth, 7 http://dx.doi.org/10.1186/1471-2393-7-9. Arenz, S., Rückerl, R., Koletzko, B., & von Kries, R. (2004). Breast-feeding and childhood obesity — A systematic review. International Journal of Obesity, 28(10), 1247–1256. Bakeman, R., & Gnisci, G. (2006). Sequential observational methods. In M. Eid, & E. Diener (Eds.), Handbook of multimethod measurement in psychology (pp. 127–140). Washington, DC: American Psychological Association. Baughcum, A.E., Powers, S.W., Johnson, S.B., Chamberlin, L.A., Deeks, C.M., Jain, A., et al. (2001). Maternal feeding practices and beliefs and their relationships to overweight in early childhood. Developmental and Behavioral Pediatrics, 22(6), 391–408. Bergmeier, H., Skouteris, H., Horwood, S., Hooley, M., & Richardson, B. (2014). Associations between child temperament, maternal feeding practices and child body mass index during the preschool years: A systematic review of the literature. Obesity Reviews, 15(1), 9–18. Blissett, J., Meyer, C., & Haycraft, E. (2006). Maternal and paternal controlling feeding practices with male and female children. Appetite, 47(2), 212–219. Chatoor, I., Getson, P., Menvielle, E., Brasseaux, C., O'Donnell, R., Rivera, Y., et al. (1997). A feeding scale for research and clinical practice to assess mother–infant interactions in the first three years of life. Infant Mental Health Journal, 18(1), 76–91. de Campora, G., Giromini, L., Larciprete, G., Li Volsi, V., & Zavattini, G.C. (2014). The impact of maternal overweight and emotion regulation on early eating behaviors. Eating Behaviors, 15(3), 403–409. DiSantis, K.I., Hodges, E.A., Johnson, S.L., & Fisher, J.O. (2011). The role of responsive feeding in overweight during infancy and toddlerhood: A systematic review. International Journal of Obesity, 35(4), 480–492. Faith, M.S., & Kerns, J. (2005). Infant and child feeding practices and childhood overweight: The role of restriction. Maternal & Child Nutrition, 1(3), 164–168. Faith, M.S., Scanlon, K.S., Birch, L.L., Francis, L.A., & Sherry, B. (2004). Parent–child feeding strategies and their relationships to child eating and weight status. Obesity Research, 12(11), 1711–1722. Grube, M., Bergmann, S., Keitel, A., Herfurth-Majstorovic, K., Wendt, V., von Klitzing, K., et al. (2013). Obese parents — Obese children? Psychological–psychiatric risk factors of parental behavior and experience for the development of obesity in children aged 0–3: Study protocol. Public Health http://dx.doi.org/10.1186/1471-2458-13-1193.
136
V. Wendt et al. / Eating Behaviors 18 (2015) 131–136
Haycraft, E.L., & Blissett, J.M. (2008). Maternal and paternal controlling feeding practices with male and female children. Obesity, 16, 1552–1558. Khandpur, N., Blaine, R.E., Orlet Fisher, J., & Davison, K.K. (2014). Fathers' child feeding practices: A review of the evidence. Appetite, 78, 110–121. Lewis, C., & Lamb, M.E. (2003). Fathers' influence on the children's development: The evidence from two-parent families. European Journal of Psychology of Education, 18(2), 211–228. Lewis, M., & Worobey, J. (2011). Mothers and toddlers lunch together. The relation between observed and reported behavior. Appetite, 56, 732–736. Lumeng, J.C., & Burke, L.M. (2006). Maternal prompts to eat, child compliance, and mother and child weight status. Journal of Pediatrics, 149(3), 330–335. Lumeng, J.C., Ozbeki, T.N., Appugliese, D.P., Kaciroti, N., Corwyn, R.F., & Bradley, R.H. (2012). Observed assertive and intrusive maternal feeding behaviors increase child adiposity. The American Journal of Clinical Nutrition, 95(3), 640–647. Mallan, K.M., Daniels, L.A., Nothard, M., Nicholson, J.M., Wilson, A., Cameron, C.M., et al. (2014). Dads at the dinner table. A cross-sectional study of Australian fathers' child feeding perceptions and practices. Appetite, 73, 40–44.
Orell-Valente, J.K., Hill, L.G., Brechwald, W.A., Dodge, K.A., Pettit, G.S., & Bates, J.E. (2007). “Just three more bites”: An observational analysis of parents' socialization of children's eating at mealtime. Appetite, 48, 37–45. Pulley, C., Galloway, A.T., Webb, R.M., & Payne, L.O. (2014). Parental child feeding practices: How do perceptions of mother, father, sibling, and self vary? Appetite, 10, 96–102. Rising, R., & Lifshitz, F. (2005). Relationship between maternal obesity and infant feedinginteractions. Nutrition Journal, 4(17) http://dx.doi.org/10.1186/1475-2891-4-17. Thompson, A.L. (2013). Intergenerational impact of maternal obesity and postnatal feeding practices on pediatric obesity. Nutrition Reviews, 71(1), 55–61. Wardle, J., Sanderson, S., Guthrie, C.A., Rapoport, L., & Plomin, R. (2002). Parental feeding styles and the intergenerational transmission of obesity risk. Obesity Research, 10(6), 453–462. World Health Organization (2000). Obesity: Preventing and managing the global epidemic (No. 894). World Health Organization. Yeung, W.J., Sandberg, J.F., Davis‐Kean, P.E., & Hofferth, S.L. (2001). Children's time with fathers in intact families. Journal of Marriage and Family, 63(1), 136–154.
Contents lists available at ScienceDirect
Eating Behaviors
Parent–child interaction during feeding or joint eating in parents of different weights Verena Wendt a,b, Sarah Bergmann a,b, Katharina Herfurth-Majstorovic a,b, Anja Keitel-Korndörfer b, Kai von Klitzing b, Annette M. Klein b,⁎ a b
Leipzig University Medical Center, IFB AdiposityDiseases, Leipzig, Germany Department of Child and Adolescent Psychiatry, Psychotherapy and Psychosomatics, University of Leipzig, Leipzig, Germany
a r t i c l e
i n f o
Article history: Received 11 December 2014 Received in revised form 2 March 2015 Accepted 29 April 2015 Available online 10 May 2015 Keywords: Parent–child interaction Child feeding Mealtime interaction Obese mother Obese father
a b s t r a c t The current study investigates parent–child interaction during feeding or during joint eating, and aimed to explore differences in feeding interactions between mothers and fathers, as well as between overweight, obese and not overweight parents. 148 mothers and 148 fathers with children aged between 7 and 47 months were observed during feeding of or joint eating with their child in the laboratory. The videotaped mother–child and father–child dyads were coded using the Chatoor Feeding Scale. This scale consists of 5 subscales: Dyadic Reciprocity, Dyadic Conflict, Talk and Distraction during Feeding, Struggle for Control, and Non-Contingency. Compared to mothers, fathers showed higher readings on the Talk and Distraction scale; in all other subscales no differences were found. The comparison between overweight, obese and not overweight mother–child dyads revealed no significant differences. Differences in father–child dyads between overweight, obese and not overweight fathers were identified in the subscale Struggle for Control: overweight fathers were marked by a higher amount of Struggle for Control than obese and not overweight fathers. Taken together, differences found in the present observational study are small to moderate, and thus the current results support extant literature demonstrating that there are no differences in feeding behaviour between mothers and fathers or between obese and non-obese parents. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction The development of children's weight is influenced by many factors. While research investigating factors that contribute to overweight or obesity has previously focused on the individual, in recent decades familial factors and parent–child-interactions have received more attention. Research on feeding is unanimous that there is a connection between certain feeding practices (e.g. restriction) or maternal decisions of what to feed (e.g. to breastfeed the child), and overweight or obesity in childhood (e.g. Arenz, Rückerl, Koletzko, & von Kries, 2004; Bergmeier, Skouteris, Horwood, Hooley, & Richardson, 2014; DiSantis, Hodges, Johnson, & Fisher, 2011; Faith & Kerns, 2005; Faith, Scanlon, Birch, Francis, & Sherry, 2004), but critical gaps remain in knowledge about differences in feeding behaviour between mothers and fathers in general, and differences between obese and non-obese parents. Although the time fathers spend interacting with their children has in general increased (Yeung, Sandberg, Davis-Kean, & Hofferth, 2001), research on the feeding practices of fathers is still scant. A review by ⁎ Corresponding author at: Department of Child and Adolescent Psychiatry, Psychotherapy and Psychosomatics, University of Leipzig, Liebigstrasse 20a, 04103 Leipzig, Germany. Tel.: +49 341 9724018; fax: +49 341 9724019. E-mail address: [email protected] (A.M. Klein).
http://dx.doi.org/10.1016/j.eatbeh.2015.04.007 1471-0153/© 2015 Elsevier Ltd. All rights reserved.
Khandpur, Blaine, Orlet Fisher, & Davison (2014) considered studies that have included fathers. They found 20 studies assessing fathers' child-feeding practices via self-reports (18 studies) or via mealtime observations (2 studies). It was shown that, compared to mothers, fathers control and limit children's food intake and food access less. Instead, pressuring the child to eat was reported as a common feeding style in fathers (Khandpur et al., 2014; Orell-Valente et al., 2007; Pulley, Galloway, Webb, & Payne, 2014). In fathers, concern that the child could become overweight is associated with higher responsibility for child feeding and monitoring of what and how much the child eats (Mallan et al., 2014). However, research findings comparing the feeding pattern of parents are not consistent. A number of studies have found no differences between mothers and fathers in terms of restrictive or pressuring feeding patterns (Blissett, Meyer, & Haycraft, 2006; Haycraft & Blissett, 2008). Although there is a strong association between maternal obesity and heightened risk of obesity in childhood, feeding differences between obese, overweight and normal-weight mothers have received comparatively little attention in the literature. Thompson (2013) gives an overview of the feeding behaviour of obese mothers. Compared to normal-weight women, overweight and obese women are less likely to breastfeed their children, and more likely to wean them earlier (Amir & Donath, 2007). Haycraft and Blissett (2008) reported that
132
V. Wendt et al. / Eating Behaviors 18 (2015) 131–136
obese mothers show more restriction and control of the quantity and quality of foods provided. Furthermore, more pressure to eat during mealtime was observed in women with a higher body mass index (BMI) (Haycraft & Blissett, 2008). However, comparisons of feeding styles between obese and non-obese mothers are partially contradictory: no differences in mother–child feeding behaviour (Lewis & Worobey, 2011; Lumeng & Burke, 2006; Lumeng et al., 2012), especially after controlling for family income (Baughcum et al., 2001), or less control by obese mothers (Wardle, Sanderson, Guthrie, Rapoport, & Plomin, 2002), have also been shown. The current study therefore investigates parent–child interactions during feeding or joint eating. The central aims of the study are to examine differences (1) between mothers and fathers, and (2) between obese, overweight and not overweight mothers and fathers. As there is either a lack of research or the results in literature are inconsistent, we investigated possible differences between mother–child and father–child feeding interactions as well as between overweight, obese and not overweight mothers and fathers, in an exploratory fashion. 2. Method 2.1. Study design and sample The current study is part of the Integrated Research and Treatment Center (IFB) AdiposityDiseases at the University Medical Centre Leipzig, and was embedded in a prospective longitudinal study investigating children's psychological and physical development (Grube et al., 2013). Families were recruited via information and flyers left at kindergartens, paediatricians' practices and facilities for children and families in the area of Leipzig, Germany. Out of a larger sample, this study included families in which both parents are involved in child care and took part in a feeding/joint eating observation with their child in the laboratory. Fathers did not have to be biologically related to the participating child but could be social fathers/ stepfathers. The sample consists of N = 148 mothers and N = 148 fathers with their children (76 girls, 51.4%) aged 7 to 47 months. The participating parents were generally highly educated; for details and further information see Table 1. Out of 148 mothers and 148 fathers, n = 44 mothers and n = 37 fathers were obese (BMI ≥ 30), n = 17 mothers and n = 32 fathers were overweight (BMI ≥ 25.00 ≤ 29.99), n = 83 mothers and n = 77 fathers were normal weight (BMI ≥ 18.50 ≤ 24.99), and n = 4 mothers and n = 2 fathers were underweight (BMI b 18.50). In this study, we combined the groups of underweight and normal-weight participants, named “not overweight” (n = 87 mothers and n = 79 fathers). For the mean BMI in the different groups see Table 1. 2.2. Procedure The families were informed about the study as well as the procedure, participated voluntarily, and each parent gave informed consent prior to their inclusion in the study. The study was approved by the ethics committee of the University of Leipzig. Parents completed a set of questionnaires at home. They were invited to the laboratory of the Department of Child and Adolescent Psychiatry, Psychotherapy and Psychosomatics, University of Leipzig. The appointment was scheduled to coincide with the child's usual mealtime. Mothers or fathers were advised to bring meals the child normally ate at that time. If mothers and fathers also eat while they feed their child at home, they were encouraged to bring food for themselves and to eat with their child in the laboratory. After a warm-up interview, the father or the mother was left alone in the room with the child. They were instructed to feed their child or eat with their child as they normally would at home. Parents were free to position their child as they wished, on their lap, in a highchair
Table 1 Sociodemographics of the t1 baseline sample. Parent characteristics
Mothers N = 148
Fathers N = 148
Age in years, M (SD) Age range BMI, M (SD) Weight group, BMIM(SD) Not overweight Overweight Obese
31.44 (4.27) 21.44–43.12 26.86 (7.52)
34.76 (6.03) 22.52–56.59 26.72 (6.12)
21.87 (1.94) 26.43 (1.34) 36.88 (5.69)
22.44 (1.82) 27.33 (1.37) 35.34 (5.24)
Educational level No degree Certificate of general or secondary educationa General qualification for university entranceb University degree
0 (0.0%) 52 (35.1%) 38 (25.7%) 58 (39.2%)
3 (2.0%) 55 (37.2%) 39 (26.4%) 51 (34.5%)
Employed at t1 Yes No Parental leave or other
89 (60.1%) 29 (19.6%) 30 (20.3%)
112 (75.7%) 29 (19.6%) 7 (4.7%)
Relationship statusc Unmarried Married (living together) Married (living apart) Divorced Other
78 (52.7%) 63 (42.6%) 2 (1.4%) 3 (2.0%) 2 (1.4%)
74 (50.0%) 63 (42.6%) 1 (0.7%) 8 (5.4%) 2 (1.4%)
Child characteristics Age in months, M (SD) Age range, months Sex Girls Boys
24.39 (11.24) 7–47 76 (51.4%) 72 (48.6%)
a The certificate of general education is an elementary school diploma, which is obtained on successful graduation from grade 9; the certificate of secondary education is obtained on successful graduation from grade 10. b This group also includes parents with the entrance qualification for a university of applied sciences. c Mothers and fathers are parenting the same child. Differences in the relationship status indicated by mothers and fathers result from e.g. fathers being divorced from an earlier marriage.
or on a normal chair at the table. The feeding interactions were videotaped for about 20 min. Each parent fed the child once. Mothers and fathers determined – according to circumstances – who fed the child first. As the child could not be fed twice successively, a further appointment was arranged for the other parent. 2.3. Measures 2.3.1. Anthropometric data The body mass index (BMI) was calculated using the weight and height of mothers and fathers, measured in the lab in a standardized way. Parents were weighed without shoes on calibrated scales. A calibrated stadiometer was used to measure height. A parental BMI below 18.50 indicates underweight, a value of 18.50 to 24.99 normal weight, of 25.00 to 29.99 overweight, and a BMI above 30 suggests that the person is obese (World Health Organization, 2000). 2.3.2. Mealtime observations The videotaped mother–child and father–child interactions during feeding and joint eating were rated using the Chatoor Feeding Scale (Chatoor et al., 1997), which was developed to assess mother–child interaction during feeding. It was designed for children aged 1 month to 3 years and consists of 26 mother-specific items and 20 child-specific items. These items are arranged under 5 subscales: Dyadic Reciprocity (quality of relatedness and affective engagement between the mother and her child; 11 maternal items and 5 child items), Dyadic Conflict (overt conflicts between mothers and their children over eating;
V. Wendt et al. / Eating Behaviors 18 (2015) 131–136
4 maternal and 8 child items), Talk and Distraction during Feeding (attempts by the dyad to engage or control each other by talking or distracting each other during feeding; 2 maternal and 2 child items), Struggle for Control (efforts or behaviour by mother and child to control feeding; 4 maternal and 3 child items), and Maternal Non-Contingency (the mother's inability to interpret child cues and respond contingently; 5 maternal and 2 child items). All items are rated on a four-point Likert scale (0–3) ranging from “none”, “a little”, “pretty much” to “very much”, and depict frequency or quality of behaviour.
2.3.3. Coding The videotaped parent–child feeding interactions were rated by two trained coders. For assessment of inter-rater reliability, 23% of the videos were coded twice. The inter-rater reliability was highest for Dyadic Reciprocity (mean rICC = .89, p b .001) and for Talk and Distraction during Feeding (mean rICC = .83, p b .001). The inter-rater reliability for the Dyadic Conflict subscale was rICC = .79 (p b .001), for the Struggle for Control subscale rICC = .77 (p b .001) and for the Maternal NonContingency subscale rICC = .71 (p b .001).
2.4. Statistical analysis Data handling and all statistical analyses were carried out using IBM SPSS Statistics (release 20). First, we conducted descriptive analyses (including Chi2 tests) in order to illustrate the feeding/joint eating interactions. Second, we calculated Pearson correlation coefficients to assess associations of the Chatoor subscales. Third, we conducted a repeated measures multivariate analysis of variance (MANOVA), with parent (mother vs. father) as the within-subjects factor and the Chatoor subscales as dependent variables followed by univariate analyses (ANOVAs). Fourth, for assessing differences between obese vs. overweight vs. not overweight mothers/ fathers, we conducted MANOVAs or MANCOVAs with the betweensubjects factor weight group, controlling for sociodemographic variables, if there were at least marginal significant differences (p b .10) between weight groups.
3. Results 3.1. Description of feeding/joint eating interactions On average, feeding or joint eating lasted 16 min (Mmothers = 15.51, SDmothers = 7.40; Mfathers = 16.28, SDfathers = 7.32). Mothers and fathers did not differ in the kind of meal they offered their children (χ2 (4, N = 296) = 7.07, p = .132). 16% of dyads ate breakfast, 16% came for lunch, 15% of mothers and fathers offered mush and 53% of parents provided a snack. In 78% of cases mothers or fathers ate together with their child (n = 115 mothers, n = 117 fathers), whereas in 22% of cases only the child ate or was fed by its mother/father during the videotaped feeding/ joint eating interaction.
133
3.2. Correlations of Chatoor subscales Pearson correlation coefficients for all Chatoor subscales were calculated separately for the mother–child and father–child dyads; see Table 2. The pattern of correlations was quite similar for mother– child and father–child dyads: the highest positive correlations were found between Dyadic Conflict, Struggle for Control, and NonContingency. In contrast, these subscales were negatively related to Dyadic Reciprocity. Moreover, there was a significant positive correlation between Dyadic Reciprocity and Talk/Distraction. 3.3. Comparison of mother–child and father–child dyads in feeding/joint eating interactions In our sample, we found a fairly high mean subscale score for Dyadic Reciprocity, whereas the mean scores for Dyadic Conflict, Struggle for Control, and Non-Contingency were low. The mean subscale score for Talk/Distraction was approximately in the middle of the possible range; see Table 3. The MANOVA with parent as within-subjects factor and the Chatoor subscales as dependent variables revealed a main effect of parent, F(5, 143) = 3.01, p = .013, η2p = .095. Univariate analyses showed that mothers and fathers differed significantly with regard to Talk/Distraction, with fathers showing more Talk/ Distraction during feeding/joint eating. Moreover, there was a trend towards significance (p = .079) in Dyadic Conflict, with fathers again showing higher scores. On all other subscales there were no differences between mothers and fathers (for M, F, p and η2p see Table 3). 3.4. Comparison of mother–child interactions during feeding/joint eating of obese, overweight and not overweight mothers Before analysing group differences, we tested whether obese, overweight and not overweight parents differed in sociodemographic variables. Obese, overweight and not overweight mothers did not differ in terms of their child's sex and age or their own age (p N .429), but did differ in their level of education, F(2, 145) = 3.21, p = .043, η2p = .042, which we therefore included as control variable. The MANCOVA with weight group (obese vs. overweight vs. not overweight) as between-subjects factor, the Chatoor subscales as dependent variables and the level of education as a covariate revealed no significant effect of weight group, F(10, 282) = 0.75, p = .681, η2p = .026. There was no significant effect for the covariate level of education either, F(5, 140) = 0.65, p = .662, η2p = .023. The univariate analyses also showed that obese, overweight and not overweight mothers did not differ significantly in any of the subscales, all p N .123 (for M, SD, F and η2p see Table 4). 3.5. Comparison of father–child interactions during feeding/joint eating of obese, overweight and not overweight fathers Obese, overweight and not overweight fathers did not differ significantly regarding their child's sex, their own age or their level of
Table 2 Pearson correlation coefficients for the Chatoor subscales of the mother–child/father–child dyads (N = 148).
1 Dyadic Reciprocity 2 Dyadic Conflict 3 Talk/Distraction 4 Struggle for Control 5 Non-Contingency Two-tailed. ⁎ p b .05. ⁎⁎ p b .01. ⁎⁎⁎ p b .001.
1
2
3
4
5
1/1
−.40⁎⁎⁎/−.32⁎⁎⁎ 1/1
.20⁎/.32⁎⁎⁎ .07/.05 1/1
−.32⁎⁎⁎/−.31⁎⁎⁎ .58⁎⁎⁎/.62⁎⁎⁎
−.20⁎/−.25⁎⁎ .59⁎⁎⁎/.49⁎⁎⁎
−.03/−.03 1/1
−.06/−.15 .56⁎⁎⁎/.55⁎⁎⁎ 1/1
134
V. Wendt et al. / Eating Behaviors 18 (2015) 131–136
Table 3 Chatoor subscale means (SD) for mothers and fathers (N = 148).
Dyadic Reciprocity (16 items) Dyadic Conflict (12 items) Talk/Distraction (4 items) Struggle for Control (7 items) Non-Contingency (7 items)
Mothers
Fathers
Univariate tests
Range of subscale score (manual)a
M (SD)
M (SD)
F
η2p
0–48 0–36 0–12 0–21 0–21
32.97 (4.10) 2.74 (3.12) 4.59 (1.89) 1.47 (1.58) 0.76 (1.13)
32.93 (4.51) 3.33 (3.58) 5.33 (2.38) 1.47 (1.61) 0.87 (1.31)
0.01 3.14+ 12.47⁎⁎ 0.00 0.67
.000 .021 .078 .000 .005
MANOVA univariate tests, two-tailed; df = 1, 147. a According to the conceptualization given in the manual, the five Chatoor subscales differ in terms of the number of items. Hence, the range of possible subscale scores differs as well. + p b .10. ⁎ p b .05. ⁎⁎ p b .01. ⁎⁎⁎ p b .01.
education (p N .154), but there was a tendency towards significance for their child's age, F(2, 145) = 2.77, p = .066, η2p = .037, which we therefore included as control variable. The MANCOVA with father weight group (obese vs. overweight vs. not overweight) as between-subjects factor, the Chatoor subscales as dependent variables and the child's age as a covariate revealed no significant effect of weight group, F(10, 282) = 1.41, p = .174, η2p = .048, whereas the age of the child was highly significant, F(5, 140) = 17.44, p b .001, η2p = .384. The univariate analyses revealed a significant effect of father weight group for the subscale Struggle for Control, F(2, 144) = 6.19, p = .003, η2p = .079. A post-hoc test (Games-Howell) revealed that father–child interactions during feeding/joint eating by overweight fathers were marked by a significantly higher amount of Struggle for Control than those of both obese and not overweight fathers, whereas those of obese and not overweight fathers did not differ from each other. For all other subscales, we found no significant differences, all p N .336 (for M, SD, F and η2p see Table 5). 4. Discussion The comparison of mothers and fathers revealed no significant differences in the type of meal they offered their child, or in the number of parents eating as well vs. the number who just fed their child. The videotaped mother–child and father–child interactions during feeding and joint eating rated with the Chatoor Feeding Scale revealed differences between both genders in Talk and Distraction during feeding/ joint eating. This behaviour was shown more in father–child dyads than mother–child dyads, with moderate effect size. That means fathers allowed more distraction during feeding while they were trying to guide feeding or joint eating by telling the child what to do or not to do and to eat. Those are specific terms of verbal prompts. However, we found no differences between mother–child and father–child dyads in most of the subscales. Father–child dyads showed equally high Dyadic Reciprocity and equally low Struggle for Control and NonContingency. This is in line with research that indicated no differences in feeding styles between mothers and fathers (Blissett et al., 2006; Haycraft & Blissett, 2008), and it may suggest that the interaction
pattern of fathers during feeding or joint eating is comparable to the behaviour of mothers. Fathers seem as able as mothers to show affective engagement and involvement during interaction, to respect the child's cues, to interpret child cues correctly, and to respond contingently. These results fit with evidence suggesting that mothers and fathers are similar in their interaction styles and that paternal patterns closely resemble maternal patterns throughout development (Lewis & Lamb, 2003). To summarize, while fathers show more verbal prompts (Talk and Distraction), they show the same degree of control on a behavioural level during feeding or joint eating as do mothers. Our study did not confirm findings that fathers are more likely to press their child to eat than mothers (Khandpur et al., 2014; Orell-Valente et al., 2007; Pulley et al., 2014). Comparisons between obese, overweight and not overweight parents showed no significant differences in maternal feeding behaviour, and showed significant difference in paternal feeding behaviour only for Struggle for Control. Overweight fathers scored higher on Struggle for Control than either obese or not overweight fathers. It is possible that overweight fathers are more concerned about their child's weight and therefore try to control feeding. This is in line with the finding by Mallan et al. (2014) that a higher level of concern in fathers is associated with greater monitoring of the child's eating. However, as the multivariate effect of weight group was not significant, this result should be interpreted with caution. Apart from this we did not find any significant differences. Therefore, we concluded that overweight and obese parents show a quality of relatedness and affective engagement during feeding/ joint eating that is comparable to that of normal weight parents. We did not observe either higher levels of Dyadic Conflict or of control in obese compared to not overweight parents. Attempts at distraction were also detected equally in parent–child dyads of overweight and obese parents compared to not overweight parents. Overweight, obese and not overweight parents also show the same level of ability to interpret child cues and to respond contingently. This is in line with data on parent feeding styles, where obese mothers are not reported to be more likely to offer food to cope with emotional distress than normal-weight mothers, or to use food as a reward, or to prompt a child to eat more than he or she wanted (Lewis & Worobey, 2011; Wardle et al., 2002). Based on our results, the assumption of intergenerational transmission
Table 4 Chatoor subscale means (SD) for obese (n = 44), overweight (n = 17) and not overweight (n = 87) mothers.
Dyadic Reciprocity (16 items) Dyadic Conflict (12 items) Talk/Distraction (4 items) Struggle for Control (7 items) Non-Contingency (7 items)
Obese mothers
Overweight mothers
Not overweight mothers
Univariate tests
Range of subscale score (manual)
M (SD)
M (SD)
M (SD)
F
η2p
0–48 0–36 0–12 0–21 0–21
32.95 (4.14) 3.39 (3.38) 4.39 (2.06) 1.77 (1.76) 1.02 (1.32)
33.71 (2.44) 1.94 (2.33) 5.06 (1.34) 1.06 (1.14) 0.41 (0.71)
32.83 (4.36) 2.56 (3.09) 4.61 (1.89) 1.40 (1.54) 0.70 (1.08)
0.33 1.31 0.96 1.48 2.13
.005 .018 .013 .020 .029
MANOVA univariate tests, two-tailed; df = 2, 144.
V. Wendt et al. / Eating Behaviors 18 (2015) 131–136
135
Table 5 Chatoor subscale means (SD) for obese (n = 37), overweight (n = 32) and not overweight (n = 79) fathers.
Dyadic Reciprocity (16 items) Dyadic Conflict (12 items) Talk/Distraction (4 items) Struggle For Control (7 items) Non-Contingency (7 items)
Obese fathers
Overweight fathers
Not overweight fathers
Univariate tests
Range of subscale score (manual)
M (SD)
M (SD)
M (SD)
F
η2p
0–48 0–36 0–12 0–21 0–21
33.43 (3.63) 2.92 (3.37) 5.73 (2.28) 1.16 (1.42) a 0.70 (1.02)
31.75 (3.93) 4.25 (3.83) 5.34 (2.36) 2.38 (2.11) b 1.22 (1.60)
33.18 (5.03) 3.15 (3.56) 5.14 (2.44) 1.25 (1.32) a 0.81 (1.28)
1.10 0.90 0.30 6.19⁎⁎ 0.96
.015 .012 .004 .079 .013
MANOVA univariate tests, two-tailed; df = 2, 144. ab Different superscripts indicate group differences found in post-hoc analyses (Games-Howell). ⁎ p b .05. ⁎⁎ p b .01. ⁎⁎⁎ p b .01.
of obesity risk via the way that obese parents feed their children has to be questioned. Wardle et al. (2002) summarize this: “The stereotype of the obese mother who uses food in non-nutritive ways, so that her child too becomes obese, is more likely to be myth than fact” (p. 459). Most studies comparing feeding patterns between obese and non-obese mothers have used questionnaires. Our results therefore complement previous observational studies of this topic (de Campora, Giromini, Larciprete, Li Volsi, & Zavattini, 2014; Lewis & Worobey, 2011; Rising & Lifshitz, 2005). There are some limitations to this study. The sample may not be representative, as families took part voluntarily. Therefore, we cannot rule out a self-selection bias. The two coders rating the parent–child feeding interactions could not be blinded with regard to the weight status of mothers and fathers. We cannot exclude that this has affected their ratings — however, as we found almost no group differences, we do not consider it very likely. Another limitation is the comparatively wide age range of the children, which was chosen in order to allow recruitment of a relatively large sample size. A further limitation refers to the method used. Although observational data have the advantage of being more objective than self-rated questionnaires, the assessment in the laboratory may have altered the families' behaviour to some degree and this may have compromised the results (Bakeman & Gnisci, 2006). In addition, we analysed parents' behaviour in only one feeding interaction and did not assess how often parents feed during the day. There is also a limitation due to the use of the Chatoor Feeding Scale (Chatoor et al., 1997), which was developed exclusively for mother–child interactions and children from 1 month to 3 years of age. In this study the scale was used for evaluating father–child interactions and with children in a broader age range (max. 47 months) as well. As there are to date no instruments that have been validated for father–child feeding interactions we chose the Chatoor Feeding Scale, which in the main suited the age group of the children. On the other hand, there are several strengths to this study. First of all, we observed parent–child interaction during mealtimes, which is a more objective way of assessing feeding behaviour than using selfreports. In addition, there are almost no studies comparing the feeding interactions of obese, overweight and not overweight parents as well as the feeding interactions of mothers and fathers. This study therefore adds important information.
5. Conclusion In our study, mothers and fathers differed in interaction during feeding or joint eating with their child only in terms of behaviour, including distraction and verbal prompts. Moreover, there were almost no differences between obese and not overweight parents. In general, all parents showed interactions with their child that were characterized by affective engagement and involvement, respect for the child's cues, correct interpretation, and contingent responses to child cues.
Role of funding sources This research is part of the Integrated Research and Treatment Centre “IFB AdiposityDiseases” and is supported by the Federal Ministry of Education and Research, Germany (BMBF, FKZ: 01EO1001). The BMBF had no role in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the decision to submit the paper for publication.
Contributors All authors made an essential contribution to the manuscript. Kai von Klitzing is the principal investigator. He designed the study with Annette M. Klein. Verena Wendt, Sarah Bergmann, Katharina Herfurth-Majstorovic and Anja Keitel were involved in coordination, recruitment and establishment and assessment of the cohort. Annette M. Klein conducted the statistical analysis. Verena Wendt and Annette M. Klein wrote the first draft of the manuscript and all authors contributed to and have approved the final manuscript.
Conflict of interest All authors declare that they have no conflicts of interest.
Acknowledgments Our thanks go to all participating parents and children, as well as to the members of our research team. This study is part of the Integrated Research and Treatment Centre “IFB AdiposityDiseases” and is supported by the Federal Ministry of Education and Research, Germany (BMBF, FKZ: 01EO1001).
References Amir, L.H., & Donath, S. (2007). A systematic review of maternal obesity and breastfeeding intention, initiation and duration. BMC Pregnancy and Childbirth, 7 http://dx.doi.org/10.1186/1471-2393-7-9. Arenz, S., Rückerl, R., Koletzko, B., & von Kries, R. (2004). Breast-feeding and childhood obesity — A systematic review. International Journal of Obesity, 28(10), 1247–1256. Bakeman, R., & Gnisci, G. (2006). Sequential observational methods. In M. Eid, & E. Diener (Eds.), Handbook of multimethod measurement in psychology (pp. 127–140). Washington, DC: American Psychological Association. Baughcum, A.E., Powers, S.W., Johnson, S.B., Chamberlin, L.A., Deeks, C.M., Jain, A., et al. (2001). Maternal feeding practices and beliefs and their relationships to overweight in early childhood. Developmental and Behavioral Pediatrics, 22(6), 391–408. Bergmeier, H., Skouteris, H., Horwood, S., Hooley, M., & Richardson, B. (2014). Associations between child temperament, maternal feeding practices and child body mass index during the preschool years: A systematic review of the literature. Obesity Reviews, 15(1), 9–18. Blissett, J., Meyer, C., & Haycraft, E. (2006). Maternal and paternal controlling feeding practices with male and female children. Appetite, 47(2), 212–219. Chatoor, I., Getson, P., Menvielle, E., Brasseaux, C., O'Donnell, R., Rivera, Y., et al. (1997). A feeding scale for research and clinical practice to assess mother–infant interactions in the first three years of life. Infant Mental Health Journal, 18(1), 76–91. de Campora, G., Giromini, L., Larciprete, G., Li Volsi, V., & Zavattini, G.C. (2014). The impact of maternal overweight and emotion regulation on early eating behaviors. Eating Behaviors, 15(3), 403–409. DiSantis, K.I., Hodges, E.A., Johnson, S.L., & Fisher, J.O. (2011). The role of responsive feeding in overweight during infancy and toddlerhood: A systematic review. International Journal of Obesity, 35(4), 480–492. Faith, M.S., & Kerns, J. (2005). Infant and child feeding practices and childhood overweight: The role of restriction. Maternal & Child Nutrition, 1(3), 164–168. Faith, M.S., Scanlon, K.S., Birch, L.L., Francis, L.A., & Sherry, B. (2004). Parent–child feeding strategies and their relationships to child eating and weight status. Obesity Research, 12(11), 1711–1722. Grube, M., Bergmann, S., Keitel, A., Herfurth-Majstorovic, K., Wendt, V., von Klitzing, K., et al. (2013). Obese parents — Obese children? Psychological–psychiatric risk factors of parental behavior and experience for the development of obesity in children aged 0–3: Study protocol. Public Health http://dx.doi.org/10.1186/1471-2458-13-1193.
136
V. Wendt et al. / Eating Behaviors 18 (2015) 131–136
Haycraft, E.L., & Blissett, J.M. (2008). Maternal and paternal controlling feeding practices with male and female children. Obesity, 16, 1552–1558. Khandpur, N., Blaine, R.E., Orlet Fisher, J., & Davison, K.K. (2014). Fathers' child feeding practices: A review of the evidence. Appetite, 78, 110–121. Lewis, C., & Lamb, M.E. (2003). Fathers' influence on the children's development: The evidence from two-parent families. European Journal of Psychology of Education, 18(2), 211–228. Lewis, M., & Worobey, J. (2011). Mothers and toddlers lunch together. The relation between observed and reported behavior. Appetite, 56, 732–736. Lumeng, J.C., & Burke, L.M. (2006). Maternal prompts to eat, child compliance, and mother and child weight status. Journal of Pediatrics, 149(3), 330–335. Lumeng, J.C., Ozbeki, T.N., Appugliese, D.P., Kaciroti, N., Corwyn, R.F., & Bradley, R.H. (2012). Observed assertive and intrusive maternal feeding behaviors increase child adiposity. The American Journal of Clinical Nutrition, 95(3), 640–647. Mallan, K.M., Daniels, L.A., Nothard, M., Nicholson, J.M., Wilson, A., Cameron, C.M., et al. (2014). Dads at the dinner table. A cross-sectional study of Australian fathers' child feeding perceptions and practices. Appetite, 73, 40–44.
Orell-Valente, J.K., Hill, L.G., Brechwald, W.A., Dodge, K.A., Pettit, G.S., & Bates, J.E. (2007). “Just three more bites”: An observational analysis of parents' socialization of children's eating at mealtime. Appetite, 48, 37–45. Pulley, C., Galloway, A.T., Webb, R.M., & Payne, L.O. (2014). Parental child feeding practices: How do perceptions of mother, father, sibling, and self vary? Appetite, 10, 96–102. Rising, R., & Lifshitz, F. (2005). Relationship between maternal obesity and infant feedinginteractions. Nutrition Journal, 4(17) http://dx.doi.org/10.1186/1475-2891-4-17. Thompson, A.L. (2013). Intergenerational impact of maternal obesity and postnatal feeding practices on pediatric obesity. Nutrition Reviews, 71(1), 55–61. Wardle, J., Sanderson, S., Guthrie, C.A., Rapoport, L., & Plomin, R. (2002). Parental feeding styles and the intergenerational transmission of obesity risk. Obesity Research, 10(6), 453–462. World Health Organization (2000). Obesity: Preventing and managing the global epidemic (No. 894). World Health Organization. Yeung, W.J., Sandberg, J.F., Davis‐Kean, P.E., & Hofferth, S.L. (2001). Children's time with fathers in intact families. Journal of Marriage and Family, 63(1), 136–154.