Exploring infant signing to enhance responsive parenting: Findings from the INSIGHT study

Abstract

Responsive parenting is a promising framework for obesity prevention, yet attempts to date have largely relied on parents accurately interpreting their child's cues. Infant signing or “baby sign language” could enhance these interventions by improving bidirectional parent–child communication during the preverbal and emerging language years. In a clinical trial testing, a responsive parenting intervention designed for obesity prevention, we pilot tested a brief intervention at age 40 weeks with a subset of participating dyads that taught the signing gesture of “all done” to improve parental recognition of satiety. In addition, we surveyed all participating mothers at child age 18 months on the use of infant signing gestures in the prior year. Two hundred twenty‐eight mothers completed the survey including 72 responsive parenting group mothers that received the signing instructions. A majority of mothers, 63.6%, reported teaching their infant signs in the prior year, and 61.4% of infants were using signs to communicate at 18 months (median signs = 2). The signs for “more” and “all done” were used by over half of study participants and were the most common signs used. Other signs related to eating or drinking were commonly used. Signing intervention group infants were more likely to use the sign for “all done” than controls (63.9% vs. 45.5%; P = 0.01), but there was no difference between groups with regard to the use of the sign for “more” (56.9% vs. 51.3%; P = 0.43). Signing is commonly used by parents of young children and holds potential to improve parental responsiveness and obesity prevention efforts.

Key messages.

• Responsive parenting is a promising approach for obesity prevention, but interventions have largely focused on improving parental recognition to and response to the child's cues during the preverbal and emerging language periods

• Infant signing and specifically signs related to feeding are used by many parents to improve bidirectional communication, but have not previously been tested as part of an intervention designed for obesity prevention.

• Signing is commonly used by parents, and an intervention component designed to communicate satiety was utilized by participants in a clinical trial.

1. INTRODUCTION

With data demonstrating that 23% of U.S. children ages 2–5 years are already overweight (Ogden, Carroll, Kit, & Flegal, 2014), research efforts have turned to prevention beginning during infancy, yet most attempts to date have been unsuccessful (Blake‐Lamb et al., 2016). Fortunately, educational strategies using a responsive parenting framework have shown promise at improving infant behaviours that contribute to later obesity and a range of other positive adaptive outcomes in children (Guttentag et al., 2014; Perez‐Escamilla, Segura‐Perez, Lott, & on behalf of the RWJF HER Expert Panel on Best Practices for Promotion Healthy Nutrition, 2017). Responsive parenting is defined as developmentally appropriate, prompt, and contingent parenting responses to a child's needs (Eshel, Daelmans, Mello, & Martines, 2006). We have reported the protective effects of responsive parenting on child body mass index (BMI) at age 3 years in an intervention directed at first time mothers initiated shortly after childbirth (Paul et al., 2018).

It is possible, however, that responsive parenting approaches aimed at obesity prevention are not being maximally utilized as nearly all of the literature and research related to responsive parenting and more specifically, responsive feeding, have focused on the parental responses to child cues and needs. This approach presumes that these infants and young children are uniformly clear in their communication, something that is certainly not true (McNally et al., 2016); there is natural variability in the developmental timing, frequency, and clarity of infants' cues (Carpenter, Nagell, & Tomasello, 1998; Crais, Douglas, & Campbell, 2004; Mundy et al., 2007; Murray & Trevarthen, 1986). Importantly, a mismatch of parental responsiveness to infant feeding cues has been hypothesized to play a role in the development of overweight (DiSantis, Hodges, Johnson, & Fisher, 2011). What has been missing from interventions aimed at the primary prevention of obesity, including those with a responsive parenting framework, is an intervention component aimed at improving the clarity of child cues including those related to hunger, satiety, and other causes of distress that could be mistaken by parents as hunger. Such an intervention would be more bidirectional than those directed largely at parents.

There has been widespread, increasing interest by parents in using systematic non‐verbal communication with their infants via gestures, also known as “infant signing” or “baby sign language” (Kwon, Vallotton, Kiegelmann, & Wilhelm, 2018; Pizer, Walters, & Meier, 2007), with a large number of books and other products available for parents to use for this purpose (“https://www.amazon.com/s/ref=nb_sb_noss_2?url=search-alias%3Daps&field-keywords=baby+sign+language, Accessed September, 20, 2018.”). The most common motivations for using infant signs by parents are understanding what their child wants, improving parent–child communication, and improving children's communication skills (Su, Vallotton, Griffore, Liu, & Wilhelm, 2013), and many of the first signs parents teach their preverbal children are related to eating and mealtime routines (Wang & Vallotton, 2016). Based on the evidence that using these signs benefits both the parent–child relationship and children's language development, with no harm to children, in 2011, the American Academy of Paediatrics recommended the practice of using infant signs as a temporary method of communication with preverbal children (Jana & Shu, 2010).

As part of the Intervention Nurses Start Infants Growing on Healthy Trajectories (INSIGHT) study, we pilot tested a brief infant signing intervention at age 40 weeks with a subset of participating dyads that taught the signing gesture of “all done” designed to reduce mother–infant discordance around infant satiety cues. In addition, at child age 18 months, we collected data on their use of all infant signing gestures in the prior year. This paper describes use of infant signing including signs related to meals, hunger, and satiety, among parents and children and assesses whether the brief intervention as part of the larger INSIGHT study was effective.

2. METHODS

2.1. Participants

This study was approved by the Human Subjects Protection Office of the Penn State College of Medicine. All participating mothers provided written informed consent. Mothers and their newborns were recruited in‐person by research staff shortly after delivery from one maternity ward (Penn State Milton S. Hershey Medical Center, Hershey, Pennsylvania). The enrolment period spanned January 2012 to March 2014. Major eligibility criteria were full‐term (≥37 weeks' gestation), singleton newborns delivered to English‐speaking, primiparous mothers ≥20 years old. Infants born <2,500 g were excluded. More details have been published elsewhere (Paul et al., 2014; Paul et al., 2018; Savage, Birch, Marini, Anzman‐Frasca, & Paul, 2016).

2.2. Intervention

The responsive parenting intervention guidance advised parents on how to respond to their child's needs across four behavioural states: drowsy, sleeping, fussy, and alert (i.e., interactive play and feeding) as has been previously described (Paul et al., 2014; Paul et al., 2018). After the trial had already started, the study team recognized the potential for infant signing to enhance responsive parenting and elected to pilot test a brief intervention component for the remaining participants randomized to the responsive parenting study group. Though the first 40‐week study visit occurred in October 2012, it was not until June 2013 that this component was approved by the Penn State Human Subjects Protection Office for inclusion in the intervention. All subsequent responsive parenting group participants (N = 84) received additional instruction on the infant sign for “all done” adapted from an instructional website (Figure 1; “All done, Finished. https://www.babysignlanguage.com/flash-cards/a/all-done/?v=7516fd43adaa, Accessed September 11, 2018.”). This instruction was given by home visiting research nurses in the context of the intervention that all responsive parenting intervention group participants received at 40 weeks advising mothers to not pressure their child to eat, how to set limits during mealtimes, and the importance of establishing consistent routines for meals and snacks. Specifically, mothers were advised to have their infant sit in a chair at a table during meal and snack times. Further, they were instructed that although mealtimes should not be negotiable, the child should be allowed to choose how much to eat from the healthy, age‐appropriate foods that are offered. The safety control group's intervention related to feeding focused on safe food temperatures, safe use of perishable foods, and choking hazards for infants and toddlers.

Figure 1.

Instructional handout given to mothers for the “all done” sign to enhance satiety responsiveness (adapted from https://www.babysignlanguage.com/flash-cards/a/all-done/?v=7516fd43adaa)

2.3. Measures

2.3.1. Background characteristics

Family demographic information was collected at enrolment. Self‐identified race/ethnicity was collected using fixed categories consistent with National Institutes of Health enrolment tables as required by the funding agency. Maternal age, pre‐pregnancy weight, and infant gestational age were extracted from medical records.

2.3.2. Infant signing

At child age 18 months, all mothers participating in INSIGHT were asked to complete an online survey related to infant signing, and 228 responded. They were asked if they had taught their child any signs within the past year, whether their child used any sign language, whether their child used the signs for “all done” and/or “more,” or any other words.

2.3.3. Feeding and appetite

At 1, 2, and 3 years, the Structure and Control in Parent Feeding questionnaire was used to assess controlling feeding practices: pressure to eat and restriction (Savage, Rollins, Kugler, Birch, & Marini, 2017). In addition, this scale assesses responsive feeding practices that may promote self‐regulation, including limiting exposure to unhealthy foods and consistent feeding routines. At age 30 months, mothers completed the Child Eating Behavior Questionnaire, measuring eating behaviours in young children (e.g., satiety and food responsiveness) (Wardle, Guthrie, Sanderson, & Rapoport, 2001), and at age 3 years, mothers completed the Child Feeding Questionnaire, a validated self‐report instrument that assesses maternal feeding beliefs and behaviours (Birch et al., 2001).

2.3.4. Anthropometrics

Trained research staff obtained all study measurements except birthweight and length, which were obtained from medical records. At the 1‐, 2‐, and 3‐year visits, measurements were performed by staff blinded to study group. At all study time points, weight was measured in duplicate to the nearest 0.1 kg using an electronic scale (Seca 354 or Seca 874; Seca). Lengths and heights were measured to the nearest 0.1 cm. At visits before age 2 years, length was measured using a recumbent length board (Shorr Productions). At 2 and 3 years, standing heights for children were measured with a stadiometer (Seca 216; Seca). Child BMI (calculated as weight in kilograms divided by height in meters squared)‐for‐age was converted to percentiles and z scores using the 2000 Centers for Disease Control and Prevention growth reference for children 2 years and older.

2.4. Statistical analyses

For analysis purposes, comparisons at child age 18 months were made between two groups, those who received the infant signing instructions and those who did not. Frequency tables were generated to describe the proportion of participants that had taught signs and were using signs in the past year including those specifically for the signs for the words “all done” and “more.” Frequency data were also calculated for the number of signs used by participating children. Comparisons between analysis groups were made using t tests, Wilcoxon rank‐sum tests, chi‐square tests, and Fisher's exact tests where appropriate without adjustment for confounders given that there were no demographic or baseline differences between study groups (Table 1). Differences in BMI z‐score, appetite, and feeding practices between those received the brief signing intervention and those who did not were assessed using analysis if variance, while controlling for INSIGHT intervention group (responsive parenting or control). Statistical significance was defined as P < 0.05, and data were analysed using SAS version 9.4.

Table 1.

Demographic characteristics of study mothers and infants by study group at 40 weeks

	Received infant signing education N = 84	No infant signing education N = 175
Infant
Male sex, no. (%)	46 (54.8)	87 (49.7)
Gestational age at birth, mean (SD), week	39.5 (1.3)	39.6 (1.1)
Mother
Age at enrolment, mean (SD), year	29.5 (4.3)	28.7 (4.8)
Hispanic ethnicity, no. (%)	6 (7.2)	11 (6.3)
Race, no. (%)
Black	6 (7.1)	7 (4.0)
White	74 (88.1)	161 (92.0)
Native Hawaiian or Pacific islander	0 (0.0)	1 (0.6)
Asian	3 (3.6)	5 (2.9)
Other	1 (1.2)	1 (0.6)
Married, no. (%)	67 (79.8)	134 (76.6)
Education no. (%)
High school graduate or less	7 (8.3)	17 (9.7)
Some college	21 (25.0)	45 (25.7)
College graduate	30 (35.7)	68 (38.9)
≥Graduate degree	26 (31.0)	45 (25.7)

3. RESULTS

Two hundred seventy‐nine (279) mother–infant dyads completed the first home visit at 3–4 weeks after birth and were considered the study cohort for the study; 259 (92.8%) dyads completed the 40‐week home visit with no significant differences between those that did and did not receive the brief infant signing intervention (Table 1). Mothers were predominantly White (90.7%), non‐Hispanic (93.4%), married (77.6%), and well educated (65.3% college graduates). At 18 months, 228 participating mothers completed the survey related to use of infant signing including 72 of the 84 responsive parenting group mothers that received the brief infant signing instructions.

A majority of mothers surveyed, 145/228 (63.6%), reported teaching their infant signs in the prior year, and 140/228 (61.4%) infants were using signs to communicate at age 18 months. Among those using signs, the median (interquartile range) number of signs used was 2 (2–4) with no significant differences between analysis groups (P = 0.95), but the median number of signs used was greater for boys than girls (3 vs. 2, P = 0.01). The maximum number of signs used by an 18‐month‐old participant was reported to be 34. The signs for “more” and “all done” were used by over half of study participants and were the most common signs used (Table 2). Several other commonly used signs related to eating or drinking.

Table 2.

Signs used by five or more study participants at age 18 months (N = 228)

Sign	Participant use N (%)
1. More	121 (53.1)
2. All done	117 (51.3)
3. Please	46 (20.1)
4. Thank you	29 (12.7)
5. Milk	28 (12.3)
6. Eat/hungry/food	27 (11.8)
7. Drink/thirsty	7 (3.1)
Dad/daddy/father	7 (3.1)
9. Mom/mommy/mother	6 (2.6)
Water	6 (2.6)
Help	6 (2.6)
12. Bath	5 (2.2)

At the 18‐month assessment, mothers who received the brief infant signing intervention at the 40‐week visit were more likely to report teaching their infant signs in the prior year than controls (75.0% vs. 58.3%, P = 0.02), and their infants were more likely to be signing at 18 months (72.2% vs. 56.4%, P = 0.02). Further, 46/72 (63.9%) of the signing intervention group infants were reported to use the sign for “all done” at age 18 months versus 71/156 (45.5%) of controls (P = 0.01). In contrast, there was no difference between groups with regard to the use of the sign for “more” (41/72 [56.9%] vs. 80/156 [51.3%]; P = 0.43).

The brief infant signing intervention was not independently associated with BMI z‐score at ages 2 or 3 years. Similarly, it was not associated with maternal feeding practices or maternal subjective ratings of child appetite.

4. DISCUSSION

This analysis of data collected from the INSIGHT study demonstrates that mothers commonly use a limited amount of signing with their infants to communicate, with the most common signs being “more,” “all done,” and others relating to eating and drinking. The high baseline rate of families using infant signs across the whole sample is notable given that the families were not selected for the study based on infant sign use. To our knowledge, this is the first study to offer such a report on rates of infant sign use. Though there was relative consistency in families' use of signs during meal times, which is similar to other reports (Wang & Vallotton, 2016), the specific contents of the signing, and the number of signs used by parents and children, was variable. Data from the current study further suggest that an intervention that incorporates signing as a means to enhance responsive parenting is feasible. Those receiving the “all done” signing instruction as part of a larger intervention focusing on recognizing hunger and satiety cues at age 40 weeks were more likely to use that sign at age 18 months.

From a broad perspective, responsive parenting is meant to foster the development of self‐regulation and promote cognitive, social, and emotional development. More recently, responsive parenting has emerged as a promising approach for obesity prevention in today's environment with encouraging findings from several trials (Daniels et al., 2012; Fangupo et al., 2015; Paul et al., 2011; Paul et al., 2018; Wen et al., 2012). Based in part on these studies and on an extensive review of the literature, the Robert Wood Johnson Foundation's Healthy Eating Research group issued guidelines in 2017 recommending responsive parenting and specifically, responsive feeding as the preferred strategy to prevent rapid infant growth and toddler overweight (Perez‐Escamilla, Segura‐Perez, & Lott, 2017). These guidelines reflected the current evidence base, however, which largely focused on parental responses to very basic infant gestures, movements, or facial expressions during the pre‐verbal and emerging language periods prior to age 2 years.

If parents are amenable to using infant signing consistently, as the results of the current study suggest, infant signing has the potential to substantially enhance responsive parenting interventions, particularly with preverbal children and those with developing verbal skills. Parents can begin using signs with their infants between ages 4 and 6 months, expecting infants to respond to and use signs between age 9 and 11 months, when they start using other intentional non‐verbal communication cues, such as waving or pointing (Fusaro & Vallotton, 2011; Vallotton, 2011a, 2011b). Children are likely to continue using infant signs until they have oral language sufficient to communicate their needs and interests (Vallotton, 2012a, 2012b) and to use them to supplement their verbal communication when they are too upset to talk (Konishi, Karsten, & Vallotton, 2018). Research on the use of infant signing has demonstrated that children who learn to sign during infancy have fewer tantrums (Acredolo & Goodwyn, 2002), better language skills as toddlers (Goodwyn & Acredolo, 2000), and improved social–emotional skills characterized by a more interactive relationship with parents and caregivers (Moore, Acredolo, & Goodwyn, 2001; Vallotton, 2012a, 2012b). Preverbal children who sign can use them to initiate communication with caregivers regarding their needs (Vallotton, 2008). Parents also report benefits for themselves including less parenting‐related stress, and more affectionate interactions, whereas observational studies have shown improved responsiveness to children's distress among experimental group parents using infant signs (Gongora & Farkas, 2009; Vallotton, 2012a, 2012b). Each of these factors could help reduce controlling feeding practices by parents, and these benefits also extend to early child education settings (Vallotton, 2011a, 2011b; Vallotton, 2009). For obesity prevention purposes, signing may allow preverbal children to indicate clearly when they are full (“all done”) and thus reduce the disparity that exists favouring parental recognition of hunger cues when compared against satiety cues (Hodges et al., 2013) particularly for those at increased multigenerational obesity risk (Gross et al., 2010).

Although the current findings did not show any associations between the brief intervention and weight, feeding, or appetite‐related outcomes, we previously reported a post hoc analysis with sex as a significant moderator of INSIGHT's responsive parenting intervention on BMI z‐score at 3 years with a larger effect for girls (Paul et al., 2018). This difference was not present at younger ages but began to emerge without reaching statistical significance at age 2 years. The current findings demonstrating greater use of signing by boys than girls at 18 months may indicate that the girls were more likely to be using oral language as their primary mode of communication by 18 months, because girls typically reach language milestones, including first words, earlier than boys (Bornstein, Hanh, & Hayes, 2004; Fenson et al., 1994; Maccoby, 1966). This raises the possibility that verbal language gaps could contribute to disparate degrees of parental responsiveness after age 18 months, something which effective signing could overcome.

This report has several limitations. The addition of a single “all done” sign is acknowledged as a small, proof‐of‐concept component to see if uptake of infant signing by parents as part of a responsive parenting intervention was possible. The population studied also was relatively homogeneous and well‐educated, so findings cannot be generalized to other populations.

In conclusion, the majority of well‐educated parents used some amount of infant signing in everyday routines with their children through age 18 months, though its use is limited and highly variable. The signs used are commonly related to eating and drinking and suggest that a signing intervention designed around responsive feeding would be of interest to many parents.

CONFLICTS OF INTEREST

The authors declare that they have no conflicts of interest.

CONTRIBUTIONS

IMP led the conception and design of the study, interpretation of the data, and drafting of the article. EEH led the data analysis and contributed to the interpretation of the data and drafting of the article. LLB coled the conception and design of the study, contributed to the interpretation of the data and critical revision of the manuscript. AS contributed to the acquisition of data and to the critical revision of the manuscript. CDV contributed to the interpretation of the data and critical revision of the manuscript. JSS contributed to the conception and design of the study, interpretation of the data, and critical revision of the manuscript. All authors approved the final version to be published.

Paul IM, Hohman EE, Birch LL, Shelly A, Vallotton CD, Savage JS. Exploring infant signing to enhance responsive parenting: Findings from the INSIGHT study. Matern Child Nutr. 2019;15:e12800 10.1111/mcn.12800