Maternal mood, body image, and eating habits predict changes in feeding practices during the COVID-19 pandemic

Abstract

Maternal mood and eating habits are associated with food parenting practices, including non-responsive feeding practices, which in turn impact children's eating habits. The COVID-19 pandemic may have negatively impacted maternal mood due to the overall stress and challenges, contributing to changes in eating behaviors and food parenting practices. The present study examined how maternal mood, body image, and eating concerns were related to perceived changes in feeding practices during the pandemic. A total of 137 mothers participated in an online study. Participants retrospectively reported their mood, eating habits, body dissatisfaction, and non-responsive feeding practices, before and during the pandemic, and responded to open-ended questions regarding changes in eating and feeding patterns during the pandemic. Results suggested differences in non-responsive feeding practices, including higher use of food as a reward for behavior and lower use of standard meal settings during the pandemic. In addition, significant relationships were found between higher maternal stress and higher body dissatisfaction (r = 0.37; p < .01), restrained eating (r = 31; p < .01), emotional eating (r = 0.44; p < .01), and higher use of overt and covert restriction retrospectively and during the pandemic. Results revealed trends in the same direction for depression and anxiety. Finally, qualitative findings were consistent with the quantitative findings, suggesting relationships between maternal mood, eating habits, and feeding practices. These results provide support for previous findings suggesting that the pandemic negatively impacted maternal well-being, increasing the use of some non-responsive feeding practices. Further work exploring the impacts of the pandemic on well-being, child feeding, and eating patterns is warranted.

1. Introduction

Prevailing conceptions of child feeding behaviors highlight how the parental environment plays a critical role (Farrow, Haycraft, & Blissett, 2015; Larsen et al., 2015; Yee, Lwin, & Ho, 2017). Parents' own trait and state attitudes and behaviors related to food and eating are understood to impact their food parenting, that is the ways in which parents interact with their children as related to food. Specifically, parental eating patterns that are disconnected from internal satiety cues, such as emotional, external, and restrained eating, have been related to child feeding practices that aim to use food instrumentally in non-responsive ways (Hoffman et al., 2014; Rodgers et al., 2014; Steinsbekk, Barker, Llewellyn, Fildes, & Wichstrøm, 2018). These parental eating patterns may be related to parental body image (Rodgers, O'Flynn, Bourdeau, & Zimmerman, 2018). During the COVID-19 pandemic, many contextual factors (e.g., stress, food access, limited shopping to name a few) may have impacted parents' own eating behaviors as well as their food parenting. The present study aimed to examine parents’ retrospective perceptions of changes in their eating and child feeding practices before and during the pandemic, and their relationship with parental anxiety, depression, and stress (see Table 4).

Table 4.

Predictors of parental feeding practices during the COVID-19 pandemic by maternal eating habit.

	RB during COVID		RE during COVID		CR during COVID		OR during COVID		SMS during COVID
	B	β	B	β	B	β	B	β	B	β
Pre-COVID	1.03**	0.911**	1.01**	0.97**	0.9**	0.91**	0.99**	0.97**	.96**	.93**
Restrained Eating	−0.11	−0.04	−0.04	1.01	0.02	0.01	−0.002	0–00	−0.02	−0.01
R2	0.821		0.95		0.82		0.94		0.86
R2 change	0.001		0–00		0–00		0–00		0–00
F	F(2,135) = .310.42**		F(2,135) = .1223.75**		F(2,135) = .315.87**		F(2,135) = .1003.52**		F(2,135) = .413.34**
Pre-COVID	1.01**	.9**	1**	.97**	.90**	.91**	1.0**	.97**	.96**	.93**
Emotional Eating	0.16*	0.08*	0.11	0.03	−0.02	−0.004	−0.03	−0.01	0–00	0–00
R2	0.82*		0.95		0.82		0.94		0.86
R2 change	0.005*		0.001		0–00		0–00		0–00
F	F(2,133) = .310.56**		F(2,133) = .1210.37**		F(2,133) = .301.39**		F(2,133) = .983.49**		F(2,133) = .401.09**
Pre-COVID	1.02**	0.9**	1.01**	.97**	.91**	.91**	1.0**	1.0**	.96**	.93**
External Eating	0.15	0.04	0.04	0.01	−0.16	−0.03	−0.19	−0.03	−0.11	−0.02
R2	0.82		0.95		0.82		0.94		0.86
R2 change	0.002		0–00		0.001		0.001		0–00
F	F(2,133) = .302.92**		F(2,133) = .1185.81**		F(2,133) = .309.13**		F(2,133) = .996.41**		F(2,133) = .403.48**

**p < .01; *p < .05.

Abbreviations: RB, Reward for Behavior; RE, Reward for Eating; CR, Covert Restriction; OR, Overt Restriction; SMS, Standard Meal Setting.

Theoretical accounts of the parental transmission of eating behaviors propose that parents may utilize child feeding practices that are consistent with their own ways of relating to food. Thus, for example, parents whose own eating habits are not grounded in responsiveness to internal cues of hunger and satiety, but rather who use food for emotional regulation, or implement food rules, may be more likely to employ child feeding practices that are similarly instrumental and non-responsive (Francis, Hofer, & Birch, 2001; McPhie, Skouteris, Daniels, & Jansen, 2014; Scaglioni, Salvioni, & Galimberti, 2008; Zarychta et al., 2019). This is problematic as such eating habits in children and adults are associated with poorer mental and physical health outcomes, including disordered eating (Brown, Parman, Rudat, & Craighead, 2012; Černelič-Bizjak & Guiné, 2021). Parental eating behaviors, including eating patterns that are responsive to emotional and external circumstances as opposed to internal cues of hunger and satiety have been shown to be related to parental child feeding practices (Francis et al., 2001; Rodgers et al., 2014). Emotional eating, that is eating in response to emotions, external eating, that is eating in response to external cues such as food availability, and restrained eating, that is deliberately limiting intake are patterns that have all been identified as related to parental feeding practices (Francis et al., 2001; McPhie et al., 2014; Morrison, Power, Nicklas, & Hughes, 2013; Rodgers et al., 2014). In particular, these parental eating patterns have been associated with parental feeding behaviors that are non-responsive and therefore do not encourage children to eat in response to their own hunger and satiety cues (Francis et al., 2001; McPhie et al., 2014). Given that childhood eating patterns may track into adulthood, understanding the relationships between parental eating patterns and their use of instrumental child feeding behaviors is important.

Body image, that is the subjective thoughts and feelings that one has about their own appearance has been described as an upstream risk factor for disordered eating behaviors such as emotional and restrained eating (Hockey, Barlow, Shiels, & Donovan, 2021; Stice & Shaw, 2002). Previous research has suggested that maternal body dissatisfaction is related to disordered eating behaviors in mothers and in turn to maternal feeding practices (Rodgers et al., 2013). Therefore, body image may be an important factor to assess in conjunction with disordered eating behavior.

A second factor that may affect child feeding behaviors is mood and stress. Indeed, parents who are experiencing anxiety, depression, or stress may be more likely to employ non-responsive child feeding practices (Rodgers et al., 2014). Consistent with this, maternal stress has been associated with higher controlled feeding patterns, although notably in this study depression was associated with lower levels of controlled feeding (Gemmill, Worotniuk, Holt, Skouteris, & Milgrom, 2013). Other studies, in contrast, have found evidence of a relationship among maternal depression and higher levels of unresponsive feeding practices (Gemmill et al., 2013; Rodgers et al., 2014). Similarly, maternal stress has been associated with the use of non-responsive feeding practices (Jones & Bryant-Waugh, 2013; Swyden et al., 2017).

The COVID-19 pandemic created a context that has been documented to lead to increased anxiety, depression, and stress in most individuals due to the health risks and the social restrictions (Cheng, Mendolia, Paloyo, Savage, & Tani, 2021; Wu et al., 2020). For parents, these effects may have been compounded by home schooling and disruptions to children's routines. Previous preliminary work has started to document the ways in which the pandemic and related restrictions may have affected parental eating and child feeding behaviors. For example, regarding the role of parental mood, in a small sample of n = 36 mothers, increased depression during the pandemic was associated with increased using of controlling feeding practices in their pre-school aged children (Giuliani & Kelly, 2021). Similarly, in another study among n = 72 parents of preschool-aged children, negative mood during COVID-19 was associated with higher levels of coercive feeding practices (Loth et al., 2022).

In terms of mood, among parents of 4–6 year olds, increased parenting stress during the onset of the pandemic was associated with increased use of instrumental parental feeding practices such as using food as a reward and for emotional regulation (Gonzalez & Ventura, 2021). Among parents of children aged 7–17 years, pandemic related distress was related to increased use of restrictive feeding practices (West, Shields, Hultstrand, Frank, & Sato, 2022). These studies provide support for the existence of a relationship between parental stress and poorer mood and the use of non-responsive feeding practices. However, to date concurrent examinations of the relationships among stress, anxiety, and depression and parental instrumental feeding practices have not been conducted. Such studies would help to better parse out the role played by different parental mood dimensions in the use of instrumental child feeding practices.

Regarding the relationships between parental eating patterns and their child feeding behaviors, preliminary work during the COVID-19 period has also documented the existence of these associations. For example, among parents of 2–12 year olds, emotional overeating was associated with high use of structure-related food parenting (Jansen et al., 2022). These initial findings provide support for further investigations into the relationships between maternal eating behaviors and child feeding behaviors during the COVID-19 pandemic.

Given these existing data, the present study aimed to examine how maternal mood and body image and eating concerns were related to perceived changes in non-responsive child feeding practices during the COVID-19 pandemic. Specifically, the first aim of this study was to explore differences in reported non-responsive child feeding practices (reward for behavior, the reward for eating, overt restriction, covert restriction, and structured meal setting) before and during the pandemic. It was hypothesized that mothers would report greater use of these feedback practices as compared to retrospective reports of before the pandemic. The second aim was to examine the relationships among maternal variables, specifically maternal mood (depression, anxiety, and stress) and body image concerns and restrained and emotional eating on the one hand, and maternal feeding practices (reward for behavior, the reward for eating, overt restriction, covert restriction, and structured meal setting). Based on previous work, it was hypothesized that greater maternal depression, anxiety, and stress would be associated with higher use of non-responsive feeding practices (hyp 1a) and increased use from before the after the pandemic (Hyp 1b). In addition, it was hypothesized that greater maternal body dissatisfaction and restrained and emotional would also be associated with higher (hyp 2a) and increased use of non-responsive feeding practices (Hyp 2b). Finally, qualitative data were explored to contextualize and explore these same questions.

2. Methods

2.1. Participants

Prior to data collection, approval was obtained by Northeastern University's Institutional Review Board. Upon participation, participants were offered the option to enter a raffle for one of 4 $50 gift cards to Amazon or Target. A total of 137 self-identified mothers participated in the study, with ages ranging from 25 to 49 (M = 37.23, SD = 7.06). Eligibility criteria included being 18 years or older and having a child 2–5 years of age. Moreover, the population was mostly White (91.2%) with other races and ethnicities also being represented in the sample, including Asian (5.1%), Hispanic or Latinx (1.5%), and Black or African American (AA) (1.5%). Additionally, 2.2% of the participants identified as being a part of another race and ethnicity, and 0.7% preferred not to label their race or ethnicity.

Participants’ highest level of education varied, with 67.2% having a graduate degree, 20.4% being a college graduate, 7.3% having completed some college, 4.4% having completed some graduate school, and 0.7% being a high school graduate. Finally, most of the participants were married (93.4%), while other marital statuses represented included being single or never married (2.2%), in a committed relationship (1.5%), divorced (1.5%), and other statuses (1.5%).

2.2. Measures

2.2.1. Depression, anxiety, and stress scale (DASS-21)

The DASS-21 is a 21-item self-report questionnaire that measures depression, anxiety, and stress. Participants indicate how closely a list of statements has applied to them over the past week, on a scale ranging from 0 (Never) to 3 (Almost Always). Some statements include: “I couldn't seem to experience any positive feelings at all” to assess for depression and “I felt I was close to panic” to assess for anxiety. Good internal reliability has been found previously (Samani & Joukar, 2007), with a Cronbach's α = 0.82 for the depression scale, α = 0.90 for the anxiety scale, and α = 0.93 for the stress scale. Similarly, in our sample, we found Cronbach's α = 0.86 for the depression scale, α = 0.73 for the anxiety scale, and α = 0.86 for the stress scale.

2.2.2. Dutch eating behavior questionnaire (DEBQ)

The DEBQ is a 33-item self-report questionnaire intended to measure three eating behaviors among adults: emotional eating (13 items), external eating (10 items), and restrained eating (10 items) (Domoff, 2015, pp. 1–2). More specifically, participants were presented with a list of questions about their eating habits since the beginning of the COVID-19 pandemic, or since the beginning of homeschooling. Examples of questions on the DEBQ include: “Do you have the desire to eat when you are irritated?” to assess for emotional eating, “If food tastes good, do you eat more than usual?” to assess for external eating, and “When you put on weight, do you eat less than you usually do?” to assess for restrained eating. The response options on the DEBQ range from 1 (Never) to 5 (Very often).

Good internal consistency for the DEBQ has been found (Van Strien, Frijters, Bergers, & Defares, 1986), with a Cronbach's α > 0.7 in all three subscales. Furthermore, internal reliability for our sample was also good, revealing a Cronbach's α = 0.96 for emotional eating, α = 0.83 for external eating, and α = 0.91 for restrained eating.

2.2.3. Body dissatisfaction subscale of the eating disorder inventory (EDI-3)

The body dissatisfaction subscale of the EDI-3 was used to assess individuals' general discontentment with the shapes and sizes of different parts of the body. The subscale includes 10 items scored on a scale ranging from 1 (Always) to 6 (Never). Statements included: “My stomach is too big” and “I feel satisfied with the shape of my body.” Higher scores indicate higher levels of body dissatisfaction. In a previous study, good internal reliability was found for the body dissatisfaction subscale (a = 0.89) among a sample of postpartum women (Rodgers, Fischer, & Zimmerman, 2022). Additionally, internal reliability for our sample was also found to be good, revealing a Cronbach's α > 0.80 for the body dissatisfaction subscale.

2.2.4. Feeding practices and structure questionnaire (FPSQ)

The FPSQ is a 40-item self-report questionnaire measuring parental feeding practices, specifically parental responsiveness to children's hunger and satiety signals. Of the original nine factors, this study utilized five non-responsive feeding practices including reward for behavior, the reward for eating, structured meal setting, overt restriction, and covert restriction. Mothers were asked to respond rate on a scale ranging from 1 (Never) to 5 (Always) first, referring to their practices before the beginning of COVID-19-related restrictions or “Usually” and second, thinking of their feeding practices since the beginning of the COVID-19 related restrictions, or “Since schools and childcare have closed.”

Jansen, Mallan, Nicholson, and Daniels (2014) found acceptable model fit and internal reliability (Cronbach's α = 0.61–0.89) for all nine factors. Moreover, our sample was found to have acceptable to good internal reliability for the following subscales: reward for behavior (α = 0.83), the reward for eating (α = 0.83), overt restriction (α = 0.67), covert restriction (α = 0.81), and structured meal setting (α = 0.86).

2.2.5. Qualitative questions

A set of open-ended questions were also provided to participants, where they were asked to describe the way the COVID-19 pandemic had impacted their family's eating and feeding, including their own eating habits, their children's eating habits, as well as their feeding practices. Participants responded to a total of three open-ended questions, including 1) What is the biggest change you have noticed to your own eating? 2) What is the biggest change you have noticed to the way you have been feeding your baby or children? And 3) What is the biggest change you have noticed to the way your children have been feeding/eating?

3. Data analysis

The IBM Statistical Package for Social Sciences (SPSS) was utilized for all quantitative data analysis. Missing data were replaced through mean imputation where appropriate. Descriptive statistics were examined for all study variables. A paired sample t-test was performed to compare differences between feeding practices retrospectively and during the COVID-19 pandemic. Relationships between feeding practices (retrospectively and during the COVID-19 pandemic), maternal body image, eating behaviors, and mood (depression, anxiety, and stress) were investigated through Pearson correlation (r) analyses. Pearson correlation (r) analyses were also performed to explore the relationships between maternal body dissatisfaction, eating habits, and feeding practices.

Multiple regression analyses were conducted to explore the relationship between mood (depression, anxiety, and stress) and non-responsive feeding practices, after controlling for retrospective feeding practices. Multiple regression analyses were also conducted to explore the relationship between maternal eating habits and non-responsive feeding practices, after controlling for retrospective feeding practices.

Thematic analysis was conducted to analyze free-text responses. Research team members, including a faculty member and a graduate-level student, developed a coding system. As a first step, the coders reviewed the responses, identifying themes and sub-themes independently. An initial codebook was developed based on the themes and sub-themes identified by the research team, which was utilized for authors to code responses separately. Upon review of the independent coding, a final codebook was developed (Braun & Clarke, 2014).

The following themes were developed and used for the analyses of this study: 1) maternal eating behaviors, 2) maternal eating disordered (ED) behaviors, 3) food shopping, preparation, and planning; 4) family meals; 5) maternal exercise, 6) maternal weight and shape, 7) types of food and availability; 8) negative affect; 9) food parenting, 10) children's eating behaviors; and 11) pregnancy.

To ensure interrater reliability, researchers randomly selected a subsample of 10 participants to be coded independently by each author. Percentage agreement was calculated following Miles and Huberman's (1994) formula where the total number of agreements is divided by the number of agreements plus the number of disagreements between coders, Results were then multiplied by 100, revealing good percentage agreement, with interrater reliability of 80%.

4. Results

4.1. Descriptive statistics

Table 1 summarizes comparisons between self-reported maternal feeding practices before and during the pandemic. Results indicated a significantly higher use of food as a reward for behavior during the pandemic (M = 6.41; SD = 2.14) compared to before (M = 6.26; SD = 1.91); t(150) = −2.03, p = .04. On the contrary, standard meal setting was found to be significantly higher before (M = 16.41; SD = 2.99) as opposed to during the pandemic (M = 16.17; SD = 3.08); t(150) = 2.64, p = .01. No other significant differences were found. Thus, only partial support was found for hypothesis 1.

Table 1.

Comparison of feeding practices before and during the pandemic.

	Pre-COVID		During COVID		t-test	p	Cohen's d
	M	SD	M	SD
Reward for Behavior	6.26	1.91	6.41	2.14	−2.03	.04*	.90
Reward for Eating	8.95	3.35	9.02	3.44	−1.12	.27	.80
Covert Restriction	11.38	3.70	11.56	3.69	−1.42	.16	1.61
Overt Restriction	13.37	3.84	13.33	3.96	.47	.63	.98
Structured Meal Setting	16.41	2.99	16.17	3.08	2.64	.01**	1.11

**p < .01; *p < .05.

4.2. Bivariate associations among variables

Correlation analysis results are summarized in Table 2 . As expected (hyp 1a), correlation analyses revealed that higher depression was associated with higher overt restriction retrospectively (r = 0.29; p < .01) and during the pandemic (r = 0.30; p < .01) as well as with higher covert restriction retrospectively (r = 0.27; p < .01) and during the pandemic (r = 0.24; p < .01). Moreover, correlation analyses also revealed a relationship between higher depression, higher body dissatisfaction (r = .30; p < .01), emotional eating (r = 0.36; p < .01), and restrained eating (r = 0.29; p < .01). No significant correlations were found between depression and reward for behavior, or reward for eating, either retrospectively or during the pandemic (see Table 3).

Table 2.

Correlation coefficients between retrospective feeding practices, feeding practices during the pandemic, maternal body image, and mood.

	Mean (SD)	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
Retrospective feeding
1.Reward for Behavior	6.26 (1.91)
2.Reward for Eating	8.95 (3.35)	.48**
3. Overt Restriction	13.37 (3.84)	.20*	.16*
4.Covert Restriction	11.37 (3.70)	.15	.03	.40**
5.Structured Meal Setting	16.41 (2.99)	−.18*	.06	−.05	.12
Feeding during the pandemic
6.Reward for Behavior	6.41 (2.15)	.91**	.40**	.21**	.15	−.15
7.Reward for Eating	9.02 (3.44)	.47**	.97**	.18*	.06	.04	.44**
8. Overt Restriction	13.33 (3.96)	.20*	.15	.97**	.40**	−.04	.22**	.19*
9.Covert Restriction	11.58 (3.69)	.13	.05	.40**	.91**	.17*	.14	.08	.41**
10.Structured Meal Setting	16.17 (3.08)	−.19*	.09	−.04	.09	.93**	−.18*	.07	−.02	.13
Maternal body image and eating
11. Body dissatisfaction	37.54 (11.02)	.04	.11	.15	.21*	−.04	.05	.10	.13	.18*	−.04
12. Restrained Eating	2.34 (.79)	.13	.07	.12	.37**	.05	.08	.06	.12	.34**	.04	.49**
13. Emotional Eating	2.68 (1.06)	.22*	.18*	.29**	.22*	−.09	.27**	.20*	.28**	.19*	−.08	.49**	.36**
Mood
14. Depression	11.06 (3.68)	.09	.01	.29**	.27**	.04	.11	.05	.30**	.24**	−.01	.30**	.29**	.36**	–
15. Anxiety	9.66 (2.57)	.14	−.04	.17*	.24**	−.02	.17*	.01	.19*	.18*	−.06	.35**	.19*	.39**	.64**
16. Stress	14.07 (4.18)	.05	−.03	.18*	.24**	−.09	.09	.02	.17*	.20*	−.13	.37**	.31**	.44**	.70**	.65**

**p < .01; *p < .05.

Table 3.

Predictors of parental feeding practices during the COVID-19 pandemic by mood.

	RB during COVID		RE during COVID		CR during COVID		OR during COVID		SMS during COVID
	B	β	B	β	B	β	B	β	B	β
Pre-COVID	1.02**	0.9**	1.004**	0.97**	0.9**	0.91**	0.99**	0.96**	0.96**	0.93**
Depression	0.12	0.03	0.3*	0.05*	−0.08	−0.01	0.14	0.02	−0.22	−0.04
R2	0.82		0.95*		0.82		0.94		0.86
R2 change	0.001		0.002*		0.00		0.00		0.001
F	F(2,133) = 300.76**		F(2,133) = 1227.41**		F(2,133) = 307.86**		F(2,133) = 988.87**		F(2,133) = 406.74**
Pre-COVID	1.02**	0.9**	1.01**	.98**	0.91**	0.92**	0.99**	0.97**	.96**	.93**
Anxiety	0.28	0.05	.49**	.05**	−0.39	−0.04	0.25	0.02	−0.4	−0.05
R2	0.82		0.95**		0.82		0.94		0.86
R2 change	0.002		0.003**		0.001		0.001		0.002
F	F(2,133) = 303.73**		F(2,133) = 1251.48**		F(2,133) = 310.74**		F(2,133) = 992.35**		F(2,133) = 409.67**
Pre-COVID	1.03**	.9**	1.01**	.98**	.91**	.91**	.99**	.97**	.96**	.92**
Stress	0.15	0.04	.28**	.05**	−0.13	−0.02	−0.01	−0.001	−0.23	−0.04
R2	0.82		0.95**		0.82		0.94		0.86
R2 change	0.002		0.002**		0.00		0.00		0.002
F	F(2,133) = 302.66**		F(2,133) = 1242.05**		F(2,133) = 308.7**		F(2,133) = 983.9**		F(2,133) = 408.52**

**p < .01; *p < .05.

Abbreviations: RB, Reward for Behavior; RE, Reward for Eating; CR, Covert Restriction; OR, Overt Restriction; SMS, Standard Meal Setting.

Higher anxiety, moreover, was found to be associated with overt restriction retrospectively (r = 0.17; p < .05) and during the pandemic (r = 0.19; p < .05) as well as with higher covert restriction retrospectively (r = 0.24; p < .01) and during the pandemic (r = 0.19; p < .05). In addition, higher anxiety was correlated with higher reward for behavior (r = 0.17; p < .05) during the pandemic, yet not retrospectively. Furthermore, higher anxiety levels were also associated with reports of higher body dissatisfaction (r = 0.35; p < .01), restrained eating (r = 19; p < .01), and emotional eating (r = 0.39; p < .01).

Similarly, higher stress was correlated with higher overt restriction retrospectively (r = 0.18; p < .05) and during the pandemic (r = 0.17; p < .05). Strong positive correlations were also found between higher stress and higher covert restriction retrospectively (r = 0.24; p < .01) and during the pandemic (r = 0.20; p < .05). On the other hand, higher stress levels were associated with reports of higher body dissatisfaction (r = 0.37; p < .01), restrained eating (r = 31; p < .01), and emotional eating (r = 0.44; p < .01). No significant correlations were found between stress and reward for behavior, reward for eating, or meal settings, either retrospectively or during the pandemic.

The relationship between maternal body dissatisfaction, eating habits, and feeding practices were also explored (Hyp 2a, Table 1). Results revealed that higher body dissatisfaction was associated with higher covert restriction retrospectively (r = 0.21; p < .05) and during the pandemic (r = 0.18; p < .05). Likewise, higher body dissatisfaction was correlated with higher maternal restrained eating (r = 0.49; p < .01) and higher maternal emotional eating (r = 0.49; p < .01). Higher maternal restrained eating was also correlated with higher covert restriction (of their child's access to food) retrospectively (r = 0.37; p < .01) and during the pandemic (r = 34.; p < .01).

Higher maternal emotional eating, moreover, was correlated with higher levels of all assessed feeding practices retrospectively except for structured meal setting, including a higher reward for behavior (r = 0.22; p < .05), reward for eating (r = 0.18; p < .05), overt restriction (r = 0.29; p < .01), and covert restriction (r = 0.22; p < .05). Similarly, higher maternal emotional eating was correlated with higher levels of all feeding practices during the pandemic except for structured meal setting, including a higher reward for behavior (r = 0.27; p < .01), reward for eating (r = 0.20; p < .05), overt restriction (r = 0.28; p < .01), and covert restriction (r = 0.19; p < .05).

4.3. Associations among maternal characteristics and child feeding

Multiple regression analysis findings revealed that, after controlling for retrospective feeding practices, mothers’ rewarding their children for eating during the pandemic was associated with higher levels of maternal depression (β = 0.042, p = .034), maternal anxiety (β = 0.052, p = .009), and maternal stress (β = 0.048, p = .015). No other significant relationships were found between maternal mood and feeding practices during the pandemic after controlling for retrospective feeding practices. Thus, only partial support was found for hyp 1b.

Moreover, multiple regression results revealed that, after controlling for retrospective feeding practices, mothers’ use of food as a reward for behavior during the pandemic was associated with higher levels of maternal emotional eating (β = 0.076, p = .044). The relationship between rewarding their children for eating feeding practices during the pandemic and maternal emotional eating revealed a trend in the same direction (β =. 033, p = .099). No other significant relationships were found between maternal eating habits and feeding practices during the pandemic after controlling for retrospective feeding practices. Thus, limited support was found for hyp 2b.

4.4. Qualitative findings

Participants described the way the environment changed during the pandemic, impacting their own eating, their children's eating habits, and their feeding practices. More specifically, mothers reported changes in food shopping, preparation, and planning (n = 82), family meals (n = 39), and types of food available (n = 86). Some mothers reported having more time and energy for prepping, planning, and cooking (n = 14) since the beginning of the pandemic. For instance, one mother reported having more time to prepare meals and to feed her son (032): “I find it easier to feed my son due to the extra time I have to put thought and effort into his meals.”

Participants also reported changes in trips to the grocery store, mostly to limit exposure to COVID-19 (n = 12). As a result, some mothers reported having restricted access to food and having to change the amount of food consumed to avoid food waste, suggesting a relationship between restrictive access to food and changes to their own eating habits. For instance, when asked about the biggest change they had noticed in their own eating habits, some mothers reported:

“I am eating less than usual as we are trying to eat what we have in our home and reduce the number of times we leave the house” (074) and;

“… In our household, we are doing fewer trips to the grocery store to decrease our exposure … This forces us to eat our produce/meat before it goes bad … it also decreases our likeliness to indulge in treats and junk food that we didn't get in our grocery run.” (015).

In addition, mothers who reported fewer trips to the grocery store oftentimes reported changes in feeding practices, including less overt restriction and variations to meal settings. For instance, the following mothers reported the ways in which exposure avoidance impacted feeding practices in their household:

“Our little one is a picky eater. So we make sure to get everything that he eats first and work around that! He'll always have fruits available. We try to get him to eat what's on our plate, but we still have the emergency frozen dinosaur chicken nuggets” (015);

“I ate less of the food my children like so they could have it when the shops were empty, but now things are back to normal” and “I'm more lax on where they sit and feed them separately from myself and my husband.” (009).

Participants also spoke about eating and feeding behaviors. More specifically, mothers reported their own eating behaviors (n = 37), with some mothers reporting eating more (n = 17), including more emotional eating (n = 10), since the beginning of the pandemic. Mothers particularly reported eating more due to emotional concerns in connection to the lack of activities available during the period of COVID-19 related restrictions. For example, mothers reported:

“Always at home so [I have a] tendency to eat out of boredom/frustration of being stuck at home with a whiney 2-year-old. I am also pregnant so the added stress of giving birth during this has made me gravitate to sweets.” (084) and;

“I normally take pleasure in (going out-going shopping-hanging out with friends-going everywhere and anywhere) cannot happen. I'm bored and every day is very similar to the last, so I am allowing myself leniency because why the hell not?” (037).

Conversely, some mothers reported having less appetite because of emotional concerns during the COVID-19-related restrictions:

“I eat less because when I am stressed in any way, I lose appetite” (038) and;

“I have a lack of appetite at times due to anxiety.” (102).

Moreover, and concurrent with quantitative findings, mothers’ responses also suggested a relationship between maternal eating habits and feeding practices. More specifically, one mother who reported being more lenient about her own eating habits also reported being less strict about the types of food and timing of feeding her child:

“Less strict about types of food/timing of eating, etc.” and “Less strict about types of food, frequency of snacking. Offering more treats as rewards for good behavior.” (022).

Similarly, mothers’ responses suggested a relationship between maternal emotional eating and changes in feeding practices, including rewarding their children for eating, and covert restriction. For example, when asked about the changes she has noticed in her own eating habits since the pandemic, one mother responded:

“Boredom is a problem … I find I'm snacking and eating more junk.” When asked about their children's eating habits, the same mother responded, “Definitely “giving in” more … My husband definitely uses food to get the children to do things/make them happy more than I do.” And “[the children] are just asking for food more …” (005).

Similarly, another mother reported about her own eating habits, “I'm snacking more and feel less in control to eat healthily.” When asked about feeding practices, this mother responded, “I'm more likely to give in to their requests for snacks and sweets because I'm distracted by other concerns and stresses” (014).

Finally, while many mothers reported changes in overall eating and feeding practices since the beginning of the pandemic, some mothers also reported that there were no noticeable changes in eating disordered behaviors (n = 10), food parenting or feeding practices (n = 18), and in their children's eating behaviors (n = 38), since the beginning of the COVID-19 related restrictions.

5. Discussion

The aim of this study was to examine the relationships among maternal mood (depression, anxiety, and stress) and body image and eating concerns on the one hand, and maternal feeding practices (reward for behavior, the reward for eating, overt restriction, covert restriction, and structured meal setting) on the other hand. Overall, maternal mood was associated with both retrospective and COVID-19 period reports of using overt and covert restriction, however, the relationship between maternal mood and using food as a reward for behavior was reported only during the COVID-19 period. Maternal body dissatisfaction and emotional eating were also associated with non-responsive feeding practices both before and during the COVID-19 period. Together with the qualitative data, these findings suggest that the pandemic was disruptive to child feeding behaviors, in particular as related to structured meals and using food as a reward for behavior.

Mothers reported using food as a reward for behavior more during the pandemic as compared to previously, and less structured meal settings. These findings align with previous suggestions that the conditions created by the pandemic, including additional work from home, might increase parental reliance on food as a reward as a means of juggling their various responsibilities (Adams, Caccavale, Smith, & Bean, 2020). Based on previous work, it was hypothesized that greater maternal depression, anxiety, and stress would be associated with higher use of non-responsive feeding practices (Hyp 1). Bivariate findings revealed that current maternal depression, anxiety, and stress, were all related to both self-reported overt and covert restriction before the pandemic, as well as during the pandemic. However, maternal anxiety was related to the use of food as a reward for behavior only during the pandemic. These findings were further support by findings from the regression analyses, suggesting that maternal depression, anxiety, and stress were associated with increased use of food as a reward for behavior during the pandemic. These findings are consistent with those of other work suggesting that increased life changes during the COVID-19 pandemic were associated with increased use of food as a reward for behavior (Wang, Devjani, Chillakanti, Dunton, & Mason, 2021). In contrast, although self-reported levels of structured meal setting were lower during the pandemic as compared to previously, this was not associated with maternal depression, anxiety, or stress. Other research has also failed to find a relationship between COVID-19 related stress and structured meal setting (Jansen et al., 2022). The qualitative findings suggested that several practical aspects related to the restrictions and their impact on food buying, meal preparation and timing may have influenced the perceived changes in structure meal setting more than maternal mood.

In addition, it was hypothesized that greater maternal body dissatisfaction and disordered eating behaviors would also be associated with higher use of non-responsive feeding practices (Hyp 2). Consistent with previous work (Rodgers et al., 2014; Steinsbekk et al., 2018), maternal emotional eating was consistently associated with higher self-reported use of the five non-responsive feeding practices both during and previous to the pandemic period. In addition, higher maternal body dissatisfaction, restrained, and emotional eating were associated with higher use of covert restriction both previous to and during the pandemic. In addition, higher maternal emotional eating was associated with increased use of food as a reward for behavior, which extends previous work regarding the relationship between maternal emotional eating and the use of non-responsive feeding practices (McPhie et al., 2014; Rodgers et al., 2014). However, these relationships were not found to vary with the pandemic. The fact that these relationships remained consistent over time suggests that the restrictions imposed by the pandemic did not substantially impact maternal body image and eating concerns in this sample, nor strengthen their relationship with the use of non-responsive feeding patterns. Given the findings related to maternal mood, it may be that feeding practices were more impacted by maternal mood and functioning broadly, than by concerns specific to body image and eating. In addition, the qualitative findings suggested that while for some mothers the social restrictions increased levels of emotional eating, for other the opposite might be true. This could explain the lack of a clear trend emerging at the group level.

The qualitative findings provide additional support for the fact that logistical constraints impacted feeding practices during the pandemic. Mothers spoke to the increased numbers of meals needing to be provided while outlets were fewer, and some foods were less available. In addition, mothers reported avoidance of shopping to lower exposure risk which also modified family eating and feeding patterns. Interestingly, the qualitative findings did suggest a relationship between maternal eating patterns and their child's feeding patterns, with those who were experiencing more difficulties in regulating their own eating reporting being less likely to maintain rules related to meals and food for their children.

Taken together these findings have a number of implications. First, they provide additional support for previous findings suggesting that the restrictions imposed by the COVID-19 pandemic and related concerns negatively impacted maternal wellbeing and increased the use of some, although not all, non-responsive feeding practices. In particular the use of food as a reward for behavior increased during this period, and was associated with increased maternal anxiety, depression and stress as well as maternal emotional eating. These findings have implications for the targeting of resources towards mothers who are experiencing increased anxiety, whether related to the specific COVID-19 pandemic, or more broadly. Second, the findings from this study suggested that disruptions to family meals in times of uncertainty and stress may be associated with increased use of non-responsive feeding practices. This has implications for the risk and benefits of widespread public measures such as home-schooling and should be considered in the future when making such decisions.

The present study included a number of limitations. The first is the unrepresentative sample that may not be more broadly generalizable, and the lack of representation of parents who did not identify as mothers. It is also important to note that we did not account for the location of our participants and certain parts of the country experienced different levels of restrictions. Future work should aim to include parents of all gender. Second, this study was a retrospective design in that parents self-reported on their attitudes and behaviors in the past before the pandemic as well as currently during the pandemic. More robust longitudinal designs would be better suited to examining these relationships. Finally, the study assessed only a range of non-responsive parental feeding practices, it would be helpful to explore feeding practices more broadly moving forward to obtain a more detailed picture of the ways in which child feeding changed during the pandemic.

Despite these limitations, this study provides valuable data regarding self-reported changes in non-responsive child feeding practices during the pandemic and the ways in which these changes were related to maternal anxiety, depression, and stress. Our findings suggest that increased use of food as a reward for behavior was associated with anxiety, depression, and stress as well as maternal emotional eating. Resources to support mothers who are experiencing anxiety and distress should be available and include content targeting child feeding behaviors. Further work, including longitudinal research, exploring the impacts of the pandemic and associated restrictions and decreased wellbeing and child feeding and eating patterns is warranted. Furthermore, continued efforts to provide support for mental health and particularly problematic eating behaviors that have been detrimentally impacted by the COVID-19 pandemic and its aftermath are needed (Linardon et al., 2022).

To conclude, despite these limitations, this study makes an important contribution to the existing literature by examining the ways in which parental feeding practices changed during the COVID-19 pandemic and highlighting the ways in which these changes were associated with maternal mood and eating patterns. These findings provide further support for the pandemic as a disruptive element as related to child feeding and food parenting, and additional research on the longer terms effects of these disruptions is warranted.

Author's contributions

RR and EZ designed the project, RR supervised the data collection and initial analyses, and drafted the introduction and discussion. IS completed the statistical data analysis and initial drafts of the methods and results. All authors read and approved the final manuscript.

Funding

No funding was acquired for this study.

Ethical statement

This study was approved by the institutional review board of the institution. All participants provided informed consent.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Abbreviations

DASS-21

Depression, Anxiety, and Stress Scale

DEBQ

Dutch Eating Behavior Questionnaire;

EDI-3

Body Dissatisfaction Subscale of the Eating Disorder Inventory

Data availability

Data will be made available on request.