Depressive Symptoms and Perceptions of Child Difficulty Are Associated with Less Responsive Feeding Behaviors in an Observational Study of Low-Income Mothers

Abstract

Background:

Maternal depressive symptoms and perceptions of child difficulty are associated with negative effects on general development and cognitive functioning in children. The study examined associations between maternal depressive symptoms, perceptions of child difficulty, and maternal feeding behaviors in a population at elevated risk for childhood obesity.

Methods:

Participants were 138 low-income black and Hispanic mothers and their children (ages 3–5) participating in an observational study of mealtimes among Head Start families. Three dinnertime observations were conducted over 2 weeks on each family and audio/videotaped for coding. Coding included eating influence attempts and other food- and nonfood-related interactions exhibited by the mother during dinner. Mothers completed questionnaires on depressive symptoms and perceptions of child difficulty. Linear regressions were conducted, examining associations between maternal depressive symptoms, perceptions of child difficulty, and coded parent feeding behaviors.

Results:

Mothers reporting higher levels of depressive symptoms used more verbal pressure to get their child to eat during meals, were more likely to discourage child independence, and less likely to enforce table manners. Mothers reporting higher perceptions of child difficulty were less likely to have nonfood-related discussions during meals and to try to get the child to eat a different food.

Conclusions:

This study is one of the first to investigate associations between maternal depression, perceptions of child difficulty, and mother's feeding behaviors during meals using observational methodology. These results may help researchers identify specific parental characteristics and feeding practices on which to intervene when developing tailored intervention programs for reducing childhood obesity.

Introduction

Mothers shape early childhood behaviors through interactions with their child.^1,2 The quality of these interactions has the potential to promote or compromise normal child development.^3–5 Mental health issues, such as depression, can lower the quality of the mother/child relationship, resulting in difficulties for the child.^3–5 Depresssed mothers exhibit less attentiveness and responsiveness to their child's needs, impeding the ability to provide a responsive and nurturing environment.^5,6 The sustained effortful behavior often required by mothers of young children can be challenging for depressed mothers, especially when tolerance for demanding behavior is needed.³

Chronic stress is often associated with depression and together both conditions are more prominent in low-income populations.^7,8 Specifically, chronic stressors such as inadequate income and housing, unemployment, and worries about parenting have been associated with higher levels of depression in low-income mothers.⁹ Given that low-income mothers are at a higher risk for chronic stress and depression, studying these populations could provide helpful insights into negative child outcomes often seen in low-income families.

Feeding practices, goal-oriented strategies parents use to control what and how much children eat, have been associated with negative child health outcomes.¹⁰ The feeding practices most commonly examined are restriction of specific foods and pressuring a child to eat.¹⁰ Restriction has been the most predictive of childhood overweight/obesity, whereas pressure to eat has been associated with lower child weight status.^10–12

A small number of studies have examined mental health issues and their associations with feeding behaviors. A recent systematic review revealed only eight studies linking maternal mental health and parent feeding and, as expected, most of these studies examined restriction and pressure to eat.¹³ Given that some depressed mothers are more likely to be intrusive in their interactions with their child,⁶ it is not surprising that studies in the systematic review supported a positive association between depression, stress, restriction, and pressure to eat.^14–16 However, other depressed mothers are more likely to be lax and withdrawn⁶; therefore, focusing on a broader range of feeding behaviors, including uninvolved or disengaged-type feeding practices, may be more productive when examining depressed mothers. One child outcome disproportionately experienced by low-income children is weight-related health problems.¹⁷

In addition to examining a restricted range of parental feeding behaviors, most of the parents in the systematic review studies reported on their own behaviors, thus, ignoring some of the common problems with self-report such as limited awareness of participants' own behaviors, social desirability, and low report of more harsh parenting.¹⁸ When directly observing parent feeding, no association between maternal depression and use of restriction was found¹⁹; however, other types of less sensitive feeding were associated with depression (i.e., use of incentives in exchange for eating and verbal and physical pressure to eat).²⁰

Maternal perceptions of child difficulty have been associated with negative parent–child interactions as well. Numerous studies have shown that when difficult child characteristics occur, parents may become overwhelmed and experience more stress in their parenting role; this makes it difficult for parents to be firm and nurturing, resulting in more negative parenting behavior.^21–23 Given these associations, it is surprising that no studies to date have examined the relationship between parental perceptions of child difficulty and parental feeding practices. We were interested in examining the associations of both maternal depressive symptoms and maternal perceptions of child difficulty with parent–child feeding interactions. Using direct observation of feeding interactions allowed us a more comprehensive examination of what and how parents were feeding their children during eating episodes. Therefore, the overall purpose of this study was to examine the associations between depressive symptoms, perceptions of child difficulty, and observed feeding behaviors in low-income families with preschoolers.

The system used to code the observations in this study differentiated between maternal attempts to influence the child's behavior during the meal (e.g., getting the child to eat, enforcing table manners) and noninfluence attempts (e.g., general conversation about the child, food, or nonfood topics). For the influence attempts, we also calculated a measure of the directiveness of verbal and nonverbal influence attempts—that is, the degree of force that mothers used to get their child to engage in a particular behavior. Since responsive feeding practices involve promoting the child's autonomy and using nondirective strategies during feeding (vs. pressuring the child to eat), the degree of forcefulness during feeding was considered a measure of nonresponsive feeding practices.²⁴ Based on the literature reviewed above, it was expected that mothers reporting depressive symptoms and perceptions of child difficulty would focus primarily on getting their children to finish the meal, engaging in more forceful verbal and nonverbal strategies in the process. Given that depression often leads to reduced positive interactions with the child and less involvement in caregiving activities, these mothers were also expected to engage in fewer nonfood-related discussions during the meal, in fewer attempts to redirect their child's eating behavior, and in fewer attempts to teach table manners and eating skills.

Methods

Participants

A total of 138 families participated in this study. These families were recruited from Head Start centers in a large urban metropolitan area in the southeast area of the United States. Head Start is a comprehensive child development program with the overall goal of increasing school readiness in young children from low-income families.

Ethnic breakdown of the mothers in this study was 42.3% black and 57.7% Hispanic. The mean age of the mothers was 32.3 years [standard deviation (SD = 7.8)]. Many mothers in this study were employed with 56.2% working. Mothers reported a wide range of education, with almost one-third having less than a high school diploma and about one-third having attended or graduated college. Children participating in the study ranged from 3 to 5 years with a mean age of 4.4 years (SD = 0.6) equally distributed across child gender. Sixty percent of the children were categorized as normal weight and just fewer than 40% were categorized as overweight or obese. A complete list of the demographic characteristics of the sample is shown in Table 1.

Table 1.

Parent and Child Characteristics (n = 138)

Parent ethnicity—Hispanic	57.7%
Parent age (SD)	32.3% (7.8)
Parent education
Sixth grade or less	9.3%
Eight grade or less	7.7%
Attended some high school	12.3%
High school graduate or GED	26.2%
Completed technical school	10.0%
Attended some college	23.8%
College graduate	10.7%
Marital status
Married	42.4%
Never married	32.0%
Widowed, separated, divorced	25.6%
Employment status, currently employed	56.2%
Child gender—female	49.3%
Child age (SD)	4.4 (0.6)
Child BMI categories
Normal (<85th percentile)	60.3%
Overweight (85th to <95th percentile)	19.1%
Obese (≥95th percentile)	20.6%

GED, general educational development; SD, standard deviation.

Procedures

Mothers were recruited for participation in this study during parent meetings at Head Start, while dropping off or picking up their child from school, and through flyers and sign-up sheets posted at the Head Start centers. Thirty Head Start centers were contacted across three Head Start districts. Mothers were told that research staff members from our study would observe their families at dinner three times over a 2-week period in their own home. Mothers were told that the purpose of the study was to better understand family interactions during mealtime and that they would be audio/videotaped during each home observation. Before the observations, consent forms were signed and confidentiality was assured. During the observations, two cameras were placed in the family eating room so that one camera would capture the mother's face and the other camera would capture mother/child interactions. Audio/videotaping began once the food was served. Most observations took place at a kitchen or dining room table; mothers sat with the children in most cases. Although other family members were sometimes present, only the mother and targeted Head Start child were coded. At the end of each observation, questionnaires (available in English and Spanish) were left for the parent to complete; the completed questionnaires were picked up at the next observation. Incentives were provided at the end of each of the observations (total of $125). The study was reviewed and approved by the Institutional Review Board at Baylor College of Medicine.

Measures

Demographic information

A demographic questionnaire was used for parent report of demographic information, including parent age, ethnicity, education, employment status, and other descriptives. Parents also reported on their children's age and gender.

Center for Epidemiologic Studies Depression Scale

The Center for Epidemiologic Studies Depression Scale CES-D is a self-report questionnaire measuring depressive symptoms in adults (20-items).²⁵ Reliability and validity of the questionnaire have been established.^25–27 Frequency of depressive symptoms during the previous week is scored ranging from 0 to 3. Previous research supports the measurement equivalency of the CES-D in low-income samples.^28,29 A total score from the CES-D was used as an indicator of depressive symptoms for each mother in the current report.

Parenting Stress Index-Short Form

The measure of maternal perceptions of child difficulty used here was a subscale of the Parenting Stress Index (PSI). The PSI-Short Form (PSI-SF) is a parent report questionnaire measuring stress experienced by a parent during interactions with his/her child (36 items).³⁰ Three subscales capture the primary components of the parent–child dyad: Parental Distress, Parent–Child Dysfunctional Interaction, and Difficult Child.³¹ Reliability and validity of the PSI-SF have been established.^30,32 The PSI-SF has been used with low-income parents.³²

Anthropometrics

Trained staff members took height and weight measurements of the children following a standard protocol to determine BMI.³³ Measurements were taken using a stadiometer and an electronic digital scale. Children wore light clothing and were asked to remove their shoes. Measurements were recorded to the nearest 0.1 cm (height) and 0.1 kg (weight). Two height and weight measures were taken and averaged. Centers for Disease Control and Prevention Reference Standards were used to generate age- and gender-specific BMI z scores.³⁴ Children were classified as normal weight (BMI ≤ 85th percentile), overweight (BMI > 85th ≤ 95th percentile), and obese (BMI > 95th percentile).

Coding of the audio/videotapes

Before the coders began coding, the audio/videotapes were transcribed in either English or Spanish based on the language used during the observation. Audio/videotapes were coded by eight trained bilingual research staff members blind to the aims of the study. Interobserver agreement was determined by having two coders code fifteen percent of the audio/videotapes. Coders were blinded to those observations used for reliability assessment. Agreement was assessed with Cohen's kappa.³⁵ Average kappa across the audio/videotapes was 0.72, which is considered acceptable. All audio/videos were coded using Noldus Observer software (Observer XT; Noldus Information Technology, Wageningen, the Netherlands).

The coding system was developed by expanding on a previous system.³⁶ A more detailed description of the coding system can be found in Power et al.³⁷ In brief, all maternal verbal and nonverbal attempts to influence the child were coded as well as the mothers' desired behavior during each event (e.g., to encourage eating, to discourage eating). Noninfluence attempts (those designed to provide information to or get information from another) were coded as well (e.g., information about the target child, information about food). The coding system includes 31 codes for maternal behavior: 26 codes for influence attempts and 5 codes for noninfluence attempts.

Coder training

Each staff member who coded the audio/videotapes completed extensive training and certification. Group sessions were conducted with training tapes, so staff members could obtain an in-depth understanding of the constructs to be rated. Staff members then individually coded gold standard audio/video training tapes and their ratings were compared with the gold standard. Coding of the videotapes was initiated once the coders showed Kappas of 0.80 or greater on the coding system. Periodic retraining was conducted during the duration of coding to protect against observer drift. Interobserver reliability was conducted periodically during the study to ensure consistency in understanding of the coding procedures.

Data Analyses

The variables for analyses were the total frequency of influence and noninfluence attempts, along with a number of proportions corresponding to the specific coding categories. Proportions were used for the main analyses because the total number of influence and noninfluence attempts varied widely across families. For the influence attempts, separate proportions were calculated for the desired behavior categories, the verbal strategies, and the nonverbal strategies. The numerators for these proportions were the frequencies that a particular code occurred and the dominator for each was the total frequency of influence attempts (verbal plus nonverbal). For example, if a mother encouraged her child to eat 10 times during the observation and the total number of influence attempts was 50, the proportion for that mother was 0.20. Scores were averaged across the three sets of observations for each family.

To reduce the number of variables for the present analyses, composite codes were calculated by summing the proportion of various codes. For the noninfluence attempts, the five composite codes were clarification statements, information about the target child, information about other people (mother and others), information about food, and information on other nonfood topics. The seven desired behavior composite codes were attempting to get the child to eat, attempting to get the child to eat a different food, attempting to get the child to stop eating, teaching eating skills, enforcing table manners, discouraging child independence, and attempting to get the child to engage in other nonfood-related behaviors.

To reduce the number of verbal and nonverbal strategy codes, separate verbal and nonverbal pressure scores were computed. These were computed by calculating a weighted average of the verbal and nonverbal strategies with the weights corresponding to the directiveness of the strategies. The weights and strategies listed in order of decreasing directiveness are presented in Table 2. To calculate the scores, the proportion of strategies for each code was multiplied by its respective weight. These products were summed and divided by the total proportion of interactions that included verbal or nonverbal strategies.

Table 2.

Verbal and Nonverbal Strategy Codes and Weights Used to Calculate the Verbal and Nonverbal Pressure Scores (See Text)

Strategies in order of decreasing directiveness	Weights
Verbal strategies
Threaten punishment or punish	7
Promise material rewards or materially reward	6
Time pressure	5
Unelaborated commands	4
Reasons/instructs	3
Praises	2
Hints/acknowledges	1
Nonverbal strategies
Physical intervention/physical punishment	8
Spoon feeds	7
Demonstrates	6
Helps	5
Serves	4
Points/motions	3
Moves self or something closer	2
Physical affection	1

The main analyses examined the correlations between each of the maternal codes and the total CES-D score and PSI Difficult Child subscale score. Linear regressions were run for each of the maternal codes where there was a significant correlation between the CES-D or PSI score and maternal behavior. In these regressions, the predictor variables were parent ethnicity, total CES-D score, and PSI Difficult Child subscale score. Child sex was not added as a predictor because it was not correlated with any of the maternal behavior measures. We chose to run linear regressions only for those maternal behaviors showing significant correlations to keep the number of analyses to a minimum and to reduce the likelihood of type I error. Parent ethnicity was included as a control variable in these analyses because it was the only demographic variable significantly correlated with some of the observational variables associated with depressive symptoms and perceptions of child difficulty. We also ran preliminary analyses to see if the results were moderated by maternal ethnicity. No significant moderation was found. All analyses for this study were conducted using SPSS (Version 23). A level of p < 0.05 was used to determine significance. A level of p < 0.10 was used to determine trends.

Results

The simple correlations between the maternal codes and the CES-D and PSI measures are shown in Table 3. As can be seen in the table, maternal depressive symptoms were positively correlated with verbal pressure and discouraging child independence and negatively correlated with attempting to get the child to eat a different food and enforcing table manners. The Difficult Child scores were negatively correlated with attempting to get the child to eat a different food and providing or getting nonfood information. Neither the CES-D nor the PSI score correlated significantly with the total frequency of influence or noninfluence attempts.

Table 3.

Pearson Correlation Between Observed Maternal Behavior Codes, Depressive Symptoms, and Parent Perceptions of Child Difficulty

Maternal codes	CES-D	PSI difficult child
Total noninfluence attempts (frequency)	0.06	−0.08
Proportion codes^a
Clarification Statements	0.05	0.05
Information about target child	0.09	0.13
Information about others	−0.04	−0.03
Information about food	0.04	0.06
Nonfood information	−0.15	−0.21^*
Total influence attempts (frequency)	0.03	0.09
Proportion codes^b
Attempting to get the child to eat	0.13	0.06
Attempting to get the child to eat a different food	−0.18^*	−0.22^*
Attempting to get the child to stop eating	0.08	0.07
Teaching eating skills	−0.02	−0.02
Enforcing table manners	−0.21^*	−0.02
Discouraging child independence	0.20^*	0.11
Attempting to get the child to engage in other nonfood-related behaviors	0.10	0.02
Weighted strategy scores
Verbal pressure	0.17^*	0.07
Nonverbal pressure	0.01	0.10

p < 0.05.

Proportion of noninfluence attempts.

Proportion of influencing attempts.

CES-D, Center for Epidemiologic Studies Depression Scale; PSI, Parenting Stress Index.

Table 4 reports the results of the regressions. Because some of the relationships identified in the correlations became nonsignificant (p < 0.10) in the regressions, we chose to report both significant and near-significant (p < 0.10) betas in these regression equations. We made this decision because we wanted to minimize type II errors in this understudied research area. The results showed that Hispanic mothers were more likely to attempt to get their children to provide nonfood-related information or to eat a different food, whereas the black mothers used greater verbal pressure during the meal and were more likely to discourage child independence. Examination of the independent effects of depressive symptoms and perceptions of child difficulty showed that mothers reporting higher levels of depressive symptoms were more likely to discourage child independence and use more verbal pressure, whereas less depressed mothers enforced more table manners. In contrast, mothers reporting higher perceptions of child difficulty were less likely to get or provide nonfood information or to try to get the child to eat a different food.

Table 4.

Regression Analyses Predicting Maternal Behavior Codes from Caregiver Ethnicity, Depressive Symptoms, and Parent Perceptions of Child Difficulty

Maternal behavior code	F(3,130)	R ²	Standardized beta parent ethnicity (1 = Hispanic; 2 = Black)	Standardized beta CES-D score	Standardized beta difficult child score
Nonfood information	4.35^**	0.09	−0.19^*	−0.07	−0.20^*
Attempting to get the child to eat a different food	6.69^***	0.13	−0.26^**	−0.10	−0.20^*
Enforcing table manners	2.36⁺	0.05	0.02	−0.23^**	0.03
Discouraging child independence	3.14^*	0.07	0.15⁺	0.17⁺	0.07
Verbal pressure	3.30^*	0.07	0.19^*	0.15⁺	0.04

p < 0.10, ^*p < 0.05, ^**p < 0.01, ^***p < 0.001.

Discussion

Using direct observation of parent–child feeding interactions allowed us to advance our understanding of the associations between parental mental health and the feeding relationship among low-income mothers with preschoolers. The overall goal was to investigate how depressive symptoms and perceptions of child difficulty may be associated with the observed feeding behaviors of low-income Head Start families. We expected that mothers with higher depressive symptoms and perceptions of child difficulty would exhibit less responsive feeding behaviors during mealtimes using more forceful strategies to get their child to eat. We also expected these mothers to engage in fewer nonfood-related discussions, fewer attempts to redirect their children's eating behavior, and fewer attempts to teach their children manners or eating skills. We found that mothers who reported more depressive symptoms were more likely to verbally pressure their children to eat during mealtimes and discourage child independence, thereby showing less responsive feeding practices. Furthermore, mothers reporting more depressive symptoms were less likely to implement table manners. Mothers who perceived their child as difficult were less likely to attempt to get their child to eat a different food or to engage in nonfood-related discussions with their child during the meal.

One possible explanation for these findings is that depressive symptoms interfere with mothers' availability for communication and affective bond with their children.² This may cause mothers to fixate on getting certain tasks completed during mealtimes such as engaging their child in eating a particular food. This may also help to explain why mothers with symptoms of depression were more likely to verbally pressure their child to eat as well as discourage child independence. In addition, depressed mothers or mothers who perceive their child as difficult may have less energy or motivation to focus on dinnertime conversation, redirecting the child's eating behavior, and/or teaching table manners. The socialization of young children's mealtime behaviors is inherent to the parenting process.²³ Encouraging manners and avoiding messes is part of this process and may be difficult for depressed mothers who are more likely to be lax, withdrawn, and have difficulties enforcing appropriate child behavior.⁶

Our findings are consistent with previous studies linking feeding practices to maternal depression and stress.^14–16 Pressure to eat, but not restriction, was associated with maternal depressive symptoms in two studies of children ranging in age from 4 to 8 years.^15,16 In a separate study of mothers with children 6 years of age, both pressuring and restrictive feeding practices were associated with maternal depression and stress.¹⁴ In all three studies, the parent-reported feeding practices of pressuring the child to eat and restrictive feeding were assessed through the Child Feeding Questionnaire.³⁸ One of these studies also included observed maternal feeding behaviors in response to novel foods, which were not associated with maternal depression.¹⁶ Given that the protocol was designed to examine feeding behaviors in response to specific foods in a controlled setting, comparing our observations from a naturalistic setting with those from a controlled setting was difficult.

Our study extends previous studies in the feeding literature by including a broad range of observational measures of mothers' feeding behaviors. As mentioned above, previous findings have been based mostly on parent-reported measures of depression, stress, and feeding.^{14–16,39,40} By including observations of feeding in our study, we eliminated the possibility that shared method variance artificially inflated correlations between depressive symptoms, stress, and feeding behaviors. Furthermore, many of the previous studies investigated a constrained range of feeding behaviors—restriction of specific foods and pressuring the child to eat—as opposed to investigating a wider range of feeding strategies, including both responsive and nonresponsive maternal feeding.^14–16 The current findings suggest that maternal depressive symptoms and perceptions of child difficulty are associated with both more forceful feeding practices and a tendency to focus on the feeding task alone and not use mealtime as an opportunity for general conversation or socialization.

Two previous studies found no relationship between feeding behaviors and maternal depression and stress.^19,41 It is interesting that both of these studies included observations—one set of observations measured only highly controlling maternal prompts such as verbal and physical control, force-feeding, and number of food offers, whereas the other included a broader range of maternal feeding behavior observations, including verbal and physical encouragements, verbal offers, general prompts, assertive prompts, and intrusiveness.^19,41 Both samples were conducted with younger children aged 15–36 months and one study observed mother–child interactions only during snack time. It may be that the effects of depression and stress on feeding behaviors may not manifest itself until children are older and management of mealtimes becomes more difficult.

It is important to emphasize that many of the maternal feeding behaviors studied here were not associated with maternal depressive symptoms or perceptions of child difficulty. This may have been because several of the behavior categories were too broad to expect a consistent directional relationship with maternal symptoms or perceptions (e.g., clarification statements, information about food, information about the target child, information about others). Other behaviors, in contrast (e.g., discouraging eating, teaching eating skills, attempting to get the child to engage in nonfood-related behaviors, nonverbal strategies), may have not yielded significant relationships because of their low frequency of occurrence and limited variance. Finally, the significant relationships identified here were small, including a number of nonsignificant trends in the regression results. Given that studies in understudied areas reporting multiple statistical tests and small effect sizes are the most likely to generate spurious findings,⁴² it is important that these findings be replicated in other samples. The small effect sizes also may indicate that maternal depressive symptoms and perceptions of child difficulty may be only two of many possible factors that are associated with maternal behaviors at mealtime.

We acknowledge some limitations to this study when interpreting the results. Our study participants were from low-income ethnically diverse black and Hispanic Head Start families living in the southeast area of the United States. Therefore, results cannot be generalized beyond our sample. Furthermore, self-report measures were used to assess depressive symptoms and perceptions of child difficulty in our study. Even though these questionnaires have been used widely, they may not provide the most comprehensive assessments of the constructs. Finally, future studies should focus on longitudinal data to better determine directionality of relationships.

Despite these limitations, the results show many associations between maternal depressive symptoms, perceptions of child difficulty, and feeding practices consistent with the interpretation that mothers experiencing significant emotional distress may focus primarily on getting their children to finish their meal, often applying considerable pressure in the process, which may lead to increased childhood obesity risk. These findings suggest that successful interventions to impact maternal feeding practices not only need to go beyond simply helping mothers understand how to engage in responsive feeding behaviors but also need to address the perceptions of child difficulty and psychological symptoms that may contribute to nonresponsive feeding practices in the first place.

Potential practical implications may include raising awareness of the prevalence of depressive symptoms in low-income families within the community. Through raising awareness of maternal depression and stress on feeding behaviors, community programs may be developed that can directly assist families. Increased awareness could mobilize healthcare professionals to pay closer attention to these factors and offer assistance accordingly. In sum, communicating this message to the community has the potential benefit of both reducing childhood obesity and assisting mothers with mental health issues.

Footnotes

Acknowledgments

This research was supported by funds from the US Department of Agriculture (USDA 2006-55215-16695 and USDA 2011-68001-30009). This work is a publication of the US Department of Agriculture (USDA/ARS) Children's Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, Houston, Texas, and has been funded, in part, with federal funds from the USDA/ARS under Cooperative Agreement No. 58-3092-5-001. The contents of this publication do not necessarily reflect the views or policies of the USDA, nor does mention of trade names, commercial products, or organizations imply endorsement from the U.S. government. The authors have no financial relationships relevant to this article to disclose.

Author Disclosure Statement

No competing financial interests exist.

References

Dubois-Comtois

, Moss

, Cyr

, Pascuzzo

. Behavior problems in middle childhood: the predictive role of maternal distress, child attachment, and mother-child interactions. J Abnorm Child Psychol, 2013; 41:1311–1324.

McCarty

, McMahon

, Conduct Problems Prevention Research G. Mediators of the relation between maternal depressive symptoms and child internalizing and disruptive behavior disorders. J Fam Psychol, 2003; 17:545–556.

Goodman

. Depression in mothers. Annu Rev Clin Psychol, 2007; 3:107–135.

Lovejoy

, Graczyk

, O'Hare

, Neuman

. Maternal depression and parenting behavior: a meta-analytic review. Clin Psychol Rev, 2000; 20:561–592.

Bernard-Bonnin

. Maternal depression and child development. Paediatr Child Health, 2004; 9:575–583.

Cummings

, Kouros

. Maternal depression and its relation to children's development and adjustment [online]. In: Tremblay

, Peters

RDeV

, Boivin

(eds), Encyclopedia on Early Childhood Development. Centre of Excellence for Early Childhood Development: Montreal, Quebec, Canada, 2009, pp. 1–6.

Wadsworth

, Santiago

. Risk and resiliency processes in ethnically diverse families in poverty. J Fam Psychol, 2008; 22:399–410.

Santiago

, Wadsworth

, Stump

. Socioeconomic status, neighborhood disadvantage, and poverty-related stress: prospective effects on psychological syndromes among diverse low-income families. J Econ Psychol, 2011; 32:218–230.

Alegria

, Shrout

, Woo

, et al. Understanding differences in past year psychiatric disorders for Latinos living in the US. Soc Sci Med, 2007; 65:214–230.

10.

Faith

, Scanlon

, Birch

, et al. Parent-child feeding strategies and their relationships to child eating and weight status. Obes Res, 2004; 12:1711–1722.

11.

Ventura

, Birch

. Does parenting affect children's eating and weight status?. Int J Behav Nutr Phys Act, 2008; 5:15.

12.

Shloim

, Edelson

, Martin

, Hetherington

. Parenting styles, feeding styles, feeding practices, and weight status in 4–12 year-old children: a systematic review of the literature. Front Psychol, 2015; 6:1849.

13.

El-Behadli

, Sharp

, Hughes

, et al. Maternal depression, stress and feeding styles: towards a framework for theory and research in child obesity. Br J Nutr, 2015; 113:S55–S71.

14.

Mitchell

, Brennan

, Hayes

, Miles

. Maternal psychosocial predictors of controlling parental feeding styles and practices. Appetite, 2009; 53:384–389.

15.

Francis

, Hofer

, Birch

. Predictors of maternal child-feeding style: maternal and child characteristics. Appetite, 2001; 37:231–243.

16.

Goulding

, Rosenblum

, Miller

, et al. Associations between maternal depressive symptoms and child feeding practices in a cross-sectional study of low-income mothers and their young children. Int J Behav Nutr Phys Act, 2014; 11:75.

17.

Wang

, Beydoun

. The obesity epidemic in the United States-gender, age, socioeconomic, racial/ethnic, and geographic characteristics: a systematic review and meta-regression analysis. Epidemiol Rev, 2007; 29:6–28.

18.

Donaldson

, Grant-Vallone

. Understanding self-report bias in organizational behavior research. J Bus Psychol, 2002; 17:245–260.

19.

Farrow

, Blissett

. Is maternal psychopathology related to obesigenic feeding practices at 1 year?. Obes Res, 2005; 13:1999–2005.

20.

Haycraft

, Farrow

, Blissett

. Maternal symptoms of depression are related to observations of controlling feeding practices in mothers of young children. J Fam Psychol, 2013; 27:159–164.

21.

Goldstein

, Harvey

, Friedman-Weieneth

. Examining subtypes of behavior problems among 3-year-old children, part III: Investigating differences in parenting practices and parenting stress. J Abnorm Child Psychol, 2007; 35:125–136.

22.

Edhborg

, Seimyr

, Lundh

, Widström

A-M

. Fussy child-difficult parenthood? Comparisons between families with a ‘depressed’ mother and non-depressed mother 2 months postpartum. J Reprod Infant Psychol, 2000; 18:225–238.

23.

Gallagher

. Does child temperament moderate the influence of parenting on adjustment?. Dev Rev, 2002; 22:623–643.

24.

Black

, Aboud

. Responsive feeding is embedded in a theoretical framework of responsive parenting. J. Nutr, 2011; 141:490–494.

25.

Radloff

. The CES-D scale a self-report depression scale for research in the general population. Appl Psychol Meas, 1977; 1:385–401.

26.

Lewinsohn

, Hoberman

, Rosenbaum

. A prospective study of risk factors for unipolar depression. J Abnorm Psychol, 1988; 97:251–264.

27.

Weissman

, Sholomskas

, Pottenger

, et al. Assessing depressive symptoms in five psychiatric populations: a validation study. Am J Epidemiol, 1977; 106:203–214.

28.

Nguyen

, Kitner-Triolo

, Evans

, Zonderman

. Factorial invariance of the CES-D in low socioeconomic status African Americans compared with a nationally representative sample. Psychiatry Res, 2004; 126:177–187.

29.

Roberts

, Vernon

. The Center for Epidemiologic Studies Depression Scale: its use in a community sample. Am J Psychiatry, 1983; 140:41–46.

30.

Abidin

. Parenting Stress Index: Professional Manual, 3rd ed. Psychological Assessment Resources: Odessa, FL, 1995.

31.

Castaldi

. Affective and Cognitive Patterns in the Mother Child Relationship During the Second Year of Life [dissertation]. University of Virginia: Charlottesville, VA, 1990.

32.

Roggman

, Moe

, Hart

, Forthun

. Family leisure and social support: relations with parenting stress and psychological well-being in head start parents. Early Child Res Q, 1994; 9:463–480.

33.

Lohman

, Roche

, Martorell

. Anthropometric Standardization Reference Manual. Human Kinetics Books: Champaign, IL, 1998.

34.

Kuczmarski

, Ogden

, Guo

, et al. 2000 CDC growth charts for the United States: methods and development. Vital Health Stat, 2002; 246:1–190.

35.

Cohen

. A coefficient of agreement for nominal scales. Educ Psychol Meas, 1960; 20:37–46.

36.

Cousins

, Power

, Olvera-Ezzell

. Mexican-American mothers' socialization strategies: effects of education, acculturation, and health locus of control. J Exp Child Psychol, 1993; 55:258–276.

37.

Power

, Hughes

, Goodell

, et al. Feeding practices of low-income mothers: how do they compare to current recommendations?. Int J Behav Nutr Phys Act, 2015; 12:34.

38.

Birch

, Fisher

, Grimm-Thomas

, et al. Confirmatory factor analysis of the Child Feeding Questionnaire: a measure of parental attitudes, beliefs and practices about child feeding and obesity proneness. Appetite, 2001; 36:201–210.

39.

Hurley

, Black

, Papas

, Caulfield

. Maternal symptoms of stress, depression, and anxiety are related to nonresponsive feeding styles in a statewide sample of WIC participants. J Nutr, 2008; 138:799–805.

40.

Tovar

, Hennessy

, Pirie

, et al. Feeding styles and child weight status among recent immigrant mother-child dyads. Int J Behav Nutr Phys Act, 2012; 9:62.

41.

Lumeng

, Ozbeki

, Appugliese

, et al. Observed assertive and intrusive maternal feeding behaviors increase child adiposity. Am J Clin Nutr, 2012; 95:640–647.

42.

Ioannidis

JPA

. Why most published research findings are false. PLoS Med, 2005; 2:e124.