Complementary Feeding Frequency and Maternal Health: Reducing Anxiety and Improving Sleep Through Biobehavioral Pathways

Introduction

Maternal anxiety and sleep disturbance are two major health risks in the first 1–2 years after childbirth. Mothers, serving as primary caregivers, experience prolonged reductions in sleep quality lasting up to six years postpartum.^1,2 Maternal anxiety is associated with worsening sleep quality and insomnia during pregnancy, and has been closely associated with both sleep disturbances and children’s future growth and developmental outcomes.³ Evidence from the Italian cohort study indicates that maternal anxiety and sleep disturbance significantly increase Attention Deficit Hyperactivity Disorder scores in children.⁴ Moreover, maternal anxiety is linked to poor child nutrition, particularly a higher risk of underweight in low-income areas of Dhaka.⁵ A previous study has shown That Sleep disturbances in infants are linked to maternal sleep quality and mental health.⁶ Moreover, the close association between maternal mental health and infant feeding behaviors reflects the inherently strong bond between mothers and their infants.⁷ Building on our team’s previous finding from 15 Chongqing districts that proper complementary feeding supports caregiver sleep,⁸ this hospital-based study in the urban core employs validated maternal sleep scales to provide more specific and actionable evidence.

Breastfeeding and complementary feeding make up most of the nutrition in early life.⁹ Complementary feeding (CF) is defined as semisolid, soft, or solid foods other than breast, formula, or animal milk.¹⁰ According to the World Health Organization (WHO), fewer than 25% of infants aged 6–23 months achieved age-appropriate standards for dietary diversity and meal frequency.¹¹ Failure to achieve age-appropriate dietary diversity and meal frequency increases the risk of infant malnutrition and heightens susceptibility to common infections and mortality during early life.¹² The WHO has issued global recommendations on complementary feeding for children aged 6–23 months, aiming to promote optimal growth and development. Guidelines include 17 standardized indicators covering both breastfeeding and complementary feeding practices.¹³ This study focuses on the complementary feeding status of children aged 6–23 months, particularly examining key indicators such as Minimum Dietary Diversity (MDD) and Minimum Acceptable Diet (MAD).¹⁴

Previous studies have shown that maternal anxiety levels have been shown to influence both the timing and approach to complementary feeding in infants.¹⁵ Such phenomena are common among postpartum mothers, who tend to place high importance on their child’s caregiving environment and overall health. Among mothers with Major Depressive Disorder, more intensive early feeding patterns have been associated with poorer sleep efficiency and increased psychological distress.¹⁶ The cognitive-behavioral model of health anxiety suggests that emotional responses to health-related information result from individuals’ interpretations or appraisals of the situation.¹⁷ Specifically, mothers’ heightened concern for their child may lead to maladaptive assumptions about the child’s health, causing them to interpret health-related information in a particularly threatening way, which in turn triggers elevated levels of anxiety.¹⁸ Moreover, policy research from Sri Lanka indicates that strengthening healthcare workers’ support for IYCF caregiving education can reduce maternal parenting stress and, in turn, maternal anxiety.¹⁹ Fluctuations in mothers’ stress and anxiety are accompanied by shifts in key hormones, and the dynamic interactions among cortisol, oxytocin, progesterone, and estrogen may mediate these associations.²⁰ Melatonin, in particular, shows pronounced circadian rhythmicity and regulates transcriptional pathways essential for daily sleep-wake patterns, making it particularly important for understanding sleep-related difficulties.^21,22 Mothers, who often assume a larger share of infant caregiving across socioeconomic contexts, may experience elevated caregiving demands that place them at risk of hormonal dysregulation and reduced secretion of sleep-supporting melatonin and lactational oxytocin.^23,24 Additionally, nighttime infant care frequently occurs under artificial light, which disrupts melatonin’s physiological functions and impairs normal circadian entrainment, further intensifying sleep disturbances associated with infant night waking.^21,25 These findings highlight the need to examine complementary feeding in the context of maternal dual health challenges.

Current research pays insufficient attention to complementary feeding practices within the In IYCF framework, with particularly limited exploration of how these practices influence maternal anxiety and sleep quality. Infants aged 6–23 months were selected for this study, as this age range aligns with the recommended application period of the IYCF complementary feeding indicators. This study hypothesizes that adherence to appropriate IYCF practices may serve a protective role against sleep disturbances and anxiety in mothers during the complementary feeding stage. Focusing on the southwestern region of China, this research aims to provide new insights into how complementary feeding may address the dual burden of maternal sleep disruptions and anxiety.

Methods

Sample and Design

Participant data were collected from the electronic health records of the Chongqing Health Center for Women and Children, a large academic medical center in the Southwestern CN. The study employed a convenience sampling strategy. Trained physician-investigators, who underwent standardized protocol training, conducted one-to-one interviews with eligible mothers to ensure consistent and reliable data collection. Data were collected for the period from December 1, 2024, to January 31, 2025. This study has been approved by the Institutional Review Board of Chongqing Maternal and Child Health Hospital (REC reference no.2024–083), and written informed consent was obtained from all participants in accordance with the Declaration of Helsinki.

Inclusion criteria for this study were i) Mothers who voluntarily participate in the study and provide written informed consent; ii) Mothers aged 18–59 years with children aged 6–23 months; iii) Mothers without severe complications or comorbidities during pregnancy; iv) Mothers able to communicate effectively to provide reliable questionnaire information. Exclusion Criteria for this study were i) Children with severe gastrointestinal diseases, such as active bleeding ulcers; ii) Cases where the mother or child does not meet the specified age requirements; iii) Cases in which communication barriers, as assessed by trained physician-investigators, prevented the collection of valid maternal responses.

Of the 646 mother/child pairs recruited, 2 infants were excluded due to incomplete IYCF feeding practice data, 2 mothers were excluded due to missing education level information, and 99 mother/child pairs were excluded due to incomplete sleep information.

Outcome

Maternal Sleep

Maternal sleep was assessed via the Pittsburgh Sleep Quality Index (PSQI) Chinese version, which is a validated tool for assessing sleep quality in the Chinese population.²⁶ The PSQI combines assessment of seven component scores (subjective sleep quality, sleep latency, sleep duration, sleep efficiency, sleep disturbances, use of sleep medication, and daytime dysfunction) into a global sleep quality score ranging from 0 to 21, a widely used tool for evaluating individual sleep quality.²⁷ Mothers completed the PSQI under the guidance of a physician through face-to-face interviews. Internal consistency for PSQI overall Cronbach’s alpha was 0.76 and the KMO was 0.72. Besides, the reliability and validity coefficients for each dimension of the PSQI were above 0.70, indicating acceptable psychometric properties. Based on previous literature research, we defined a total PSQI score > 5 as indicative of sleep disturbance and a score ≤ 5 as indicative of good sleep quality.²⁶

Maternal Anxiety

Maternal anxiety scores were assessed via the 7-item Generalized Anxiety Disorder Questionnaire (GAD-7). Due to its diagnostic reliability, factorial, construct, and criterion validity, the GAD-7 is today among the most widely used anxiety measures, both in clinical practice and research.²⁸ Mothers completed the GAD-7 scale through face-to-face interviews conducted under the guidance of a physician. The GAD-7 assesses how often individuals have experienced seven core symptoms of generalized anxiety disorder over the past two weeks, with response options including “not at all”, “on some days”, “on more than half of the days”, and “almost every day” (scored 0–3), yielding a total score ranging from 0 to 21.²⁹ In this study, the internal consistency of the GAD-7 was acceptable, with a Cronbach’s alpha of 0.90, and the KMO measure of sampling adequacy was 0.91. Based on previous literature research, we defined a total GAD-7 score >5 as indicative of anxiety and a score ≤ 5 as indicative of no anxiety symptoms.^30,31

Exposure

Mothers were asked to recall the types of food and drink and the number of times that their children consumed them in the past 24 h prior to the survey using a food frequency questionnaire.³² IYCF includes two parts, Breastfeeding and Complementary Feeding indicators. A total of 17 IYCF indicators recommended by WHO/ the United Nations agency for Children’s Fund (UNICEF).¹⁴ In this study, we focused on the Complementary Feeding indicators section of IYCF (IYCF-CF). We selected 5 indicators [including Minimum dietary diversity (MDD), Minimum meal frequency (MMF), Minimum acceptable diet (MAD), Egg and/or flesh food consumption 6–23 months, Zero-vegetable or fruit consumption 6–23 months (ZVF)] that assess IYCF practices for infants aged 6–23 months indicators based on the age characteristics of the study subjects. Definitions for each of the 5 selected IYCF indicators are provided in Appendix Table 1 from Infant and Young Child Feeding guidance by WHO/ UNICEF.¹³

Covariates

Relevant covariates were selected using Directed Acyclic Graphs (DAGs) of the proposed relationships between potential covariates, IYCF-CF and maternal sleep, and maternal anxiety based on subject matter knowledge. Selected covariates related to the infant, the mother, and the household were included in the statistical analysis. Infant characteristics included age in months and sex. Maternal characteristics included mother’s age, job, education level, family annual income level and family foster model, and marriage status.

Analyses

Continuous variables were reported as mean ± standard deviation (SD) for normally distributed data, and as median with interquartile range [M (Q1, Q3)] for non-normally distributed data. Categorical variables were presented as counts and percentages [n (%)]. Differences in demographic, socioeconomic, maternal PSQI and maternal anxiety scores parameters between infants’ age in months were assessed using the Chi-square test of association and the t-test, for categorical and continuous data, respectively. Binary Logistic regression analysis was conducted to assess the association between IYCF-CF practices and maternal sleep quality, anxiety scores. Additionally, ordinal logistic regression models were employed to examine the associations between IYCF-CF and each of the seven PSQI components.

Sensitivity analyses were conducted by fitting three models for each regression analysis. Variables included in the models were selected based on the DAGs framework and the results of LASSO regression. Model 1 was unadjusted. Model 2 was adjusted for key demographic variables, including maternal age, occupation, education level, annual household income, and family caregiving model. Model 3 further included infant age to explore the interaction between IYCF practice and infant age in months. All statistical analyses were conducted using STATA software (version 18; Stata Corp LLC, Cary, NC, USA), and the crude odds ratio (OR) or adjusted OR (AOR) with its respective 95% confidence intervals (CI) were reported. All p-values < 0.05 were considered statistically significant.

Results

Overall Characteristics of the Overall Sample

Of 543 mother/child pairs included in the study, 49.2% of children were girls. The majority of mothers (86.9%) were ≤ 35 years old and 98.0% in a state of marriage. Mothers were primarily identified as employees of enterprises (37.8%) or housewives (37.1%). Our sample was highly educated, 91.0% of mothers reported receiving a college degree or above. 35.7% of the family’s annual income at middle class. 72.0% of children were raised by a combination of parents and grandparents. 18.6% of mothers have anxiety status, 27.1% of mothers have sleep disturbance. The distribution of maternal sleep quality was different among infants of different ages (p<0.05). See Table 1 for full participant demographic information.

Table 1 Demographic Statistics by Infant Age

Prevalence of IYCF-CF Practices Within Each Infant Age Group

Figure 1 shows that the overall prevalence of MDD, MMF, ZVF and EFF ranging from nearly 30 to 60%, yet the prevalence of MAD was lowest, ranging nearly 12%. The rates of achieving MDD, MMF, MAD and ZVF were higher in infants aged 12-23 months than in infants aged 6-11 months, while the rates of achieving EFF were opposite.

Figure 1 Attainment of overall IYCF-CF practices (A) and by infant’s age in months (B).

Prevalence of Practices of IYCF-CF by Maternal Status

As shown in Figure 2, the prevalence of MDD, MMF, ZVF and EFF ranged from nearly 24 to 65%, yet the prevalence of MAD was lowest, ranging from nearly 8 to 14%. In addition, mothers with sleep disturbance had higher rates of reaching targets for MAD, and EFF than those with good sleep quality, except for MDD, MMF and ZVF. Mothers with anxiety had higher rates of reaching targets for EFF than those with no anxiety symptoms, except for the other indicators.

Figure 2 Attainment of IYCF-CF Practices by Maternal Sleep Status (A) and Maternal Anxiety Status (B).

Relationships Between Infant Age and Maternal Sleep and Anxiety Status

As shown in Table 2, Mothers of infants aged 12–23 months had improved sleep compared with those of infants aged 6–11 months (Model 1: OR = 0.62, 95% CI: 0.42–0.92, p = 0.017; Model 2: OR = 0.61, 95% CI: 0.41–0.91, p = 0.016). No significant association was found between infant age (6–11 vs 12–23 months) and maternal anxiety in model 1 and model 2.

Table 2 Association Between Infant Age and Maternal Status

Relationships Between IYCF-CF Practices and Maternal Sleep and Anxiety Status

Table 3 presents the results that, in both Model 1 and Model 2, Mothers of children achieving MMF had significantly lower PSQI scores (OR= 0.66, 95% CI: 0.45 to 0.96, p = 0.030; OR= 0.67, 95% CI: 0.45 to 0.99, p = 0.045, respectively). No associations were found between other IYCF-CF practices and maternal sleep (p > 0.05). In model 3, MMF was significantly associated with maternal anxiety (OR= 0.53, 95% CI: 0.31 to 0.91, p= 0.021). No associations were found between other IYCF-CF practices and maternal anxiety (p > 0.05).

Table 3 Association Between IYCF-CF Indicators and Maternal PSQI and Anxiety Status

Children who achieved MMF, ZVF, and MDD were negatively associated with maternal daytime dysfunction. In addition, MMF and MDD were significantly negatively correlated with maternal sleep latency, and MMF was significantly negatively correlated with maternal subjective sleep quality. However, after adjusting for the infant age, these associations were no longer statistically significant (see Appendix Table 2).

Stratified Analyses of Interaction Effects Between IYCF-CF Practices and Infant Age

As shown in Figures 3 and 4, compared to mothers of infants aged 6–11 months with suboptimal MMF or ZVF, mothers of infants aged 12–23 months exhibited significantly better sleep quality. However, no significant interactions were observed between other IYCF-CF practices and infant age. Furthermore, the combined effect of infant age and IYCF-CF practices did not significantly influence maternal anxiety levels.

Figure 3 The joint effect of infant age and IYCF-CF practice on maternal sleep. Abbreviations: MMF, Minimum Meal Frequency; MDD, Minimum Dietary Diversity; MAD, Minimum Acceptable Diet; EFF, Egg and/or Flesh Food; ZVF, Zero Vegetable or Fruit. Notes: Demographic variables were adjusted for mother’s age, job, education level, family annual income level and family foster model.

Figure 4 The joint effect of infant age and IYCF-CF practice on maternal anxiety. Abbreviations: MMF, Minimum Meal Frequency; MDD, Minimum Dietary Diversity; MAD, Minimum Acceptable Diet; EFF, Egg and/or Flesh Food; ZVF, Zero Vegetable or Fruit. Notes: Demographic variables were adjusted for mother’s age, job, education level, family annual income level and family foster model.

Discussion

Maternal sleep disturbances and anxiety are common throughout the CF practices period and pose significant public health concerns.^4,33 A meta-analysis covering 34 countries across six continents estimated the prevalence of perinatal anxiety at nearly 20%.³⁴ Another review reported postpartum sleep disturbance rates as high as 67.2%, with anxiety prevalence ranging from 18.3% to 29.4%.³⁵ Poor maternal sleep, particularly in late pregnancy, has been associated with adverse neuropsychological and behavioral outcomes in children³⁶ and may increase the risk of psychological distress within young families.³⁷

Our study, conducted in Chongqing major urban center in southwest China-found that 27.1% of mothers experienced postpartum sleep disturbances, while 18.6% reported anxiety symptoms. Sleep disturbances were more common among mothers of infants aged 6–11 months compared to those with children aged 12–23 months.

Compared to reports from lower-income regions such as parts of the Middle East, where maternal sleep disturbance rates can reach 72%,³⁸ our findings suggest a relatively lower prevalence in Chongqing. This may be attributed to higher socioeconomic status (SES), with 53.7% of participating families reporting middle-to-upper income levels. Access to resources such as postpartum caregivers (eg, “yuesao”) may buffer sleep problems but may not mitigate psychological stress, as economic demands and societal expectations may contribute to anxiety.

Additionally, 72.6% of infants were cared for through mixed caregiving (parents and grandparents). While grandparental support can ease maternal workload, intergenerational caregiving conflicts may increase maternal stress. These findings underscore the need for targeted CF period mental health interventions that address both sleep and anxiety, particularly in rapidly urbanizing settings.

Breastfeeding and complementary feeding are key to early-life nutrition. Undernutrition contributes to 45% of infant deaths globally, with 2.7 million deaths annually in LMICs.⁹ Our study in Southwest China revealed suboptimal IYCF practices: MMF, MDD, ZVF, and EFF ranged from 30–60%, while MAD was as low as 12.5%, much lower than previous studies in urban or welfare-covered regions. In the study by other scholars in Chongqing, the proportion of non-vegetables at 7–9 months of age was close to 30%, and the proportion of egg intake was 40%.³⁹ For example, in urban Chinese ECE centers, the prevalence of MDD, MMF, and MAD was 59.4%, 60.6%, and 39.2%,⁴⁰ respectively; in rural Qinghai, these were 63.4%, 55.6%, and 39.3%.⁴¹ This may reflect that most mothers in our study were from nutrition-improvement areas not covered by national welfare (eg, Yingyangbao). Welfare support, such as WIC, is known to positively impact feeding behaviors, frequent feeding to reassure infants and young children is the mainstay.⁴² In addition, the qualified rates of MDD, MAD, and MMF were 69.2%, 68%, and 41.9%, respectively, which were significantly higher than those in other areas.⁴³ Conversely, employment and higher maternal education may paradoxically reduce feeding quality due to time constraints and stress.⁴⁴ This may also explain the low IYCF-CF compliance rate among our 68% of working women and 91.0% of those with post-college education. External factors-advertising, social norms-and cultural influences (eg, food aversions, parental overcontrol) further contribute.⁴⁵ Precise feeding often limits dietary diversity and frequency.⁴⁶ Evidence from the Sleep SAAF RP intervention⁴⁷ shows that targeted support can improve practices, suggesting that enhancing hospital-based IYCF education is essential, particularly for mothers and primary caregivers.

Maternal perceptions of sleep disturbances and stressors vary cross‑culturally, yet concerns regarding infant health and behavior emerge as a key factor connecting caregiving stress to maternal sleep and anxiety. This pattern is supported by studies from Iran, Italy, and Jordan, where infant‑related worries were identified as a major source of parental stress.^24,48,49 A qualitative study in the Congo found that maternal stress and perceived feeding concerns-such as weight loss, further exacerbated anxiety.³³ Beyond anxiety, maternal sleep disorders also pose serious health threats. Feeding infants while fatigued-often leading to mothers falling asleep during breastfeeding common and heightens the risk of accidental harm. Our results indicate a clear behavioral distinction: meeting the MMF was directly associated with a lower likelihood of maternal sleep disturbances, whereas failing to meet MMF was associated with poorer sleep outcomes. However, the presence or absence of adequate complementary feeding practices, including MMF, did not correspond to meaningful differences in maternal anxiety levels. Prior research also found no strong association between feeding practices and maternal worry.⁵⁰ From a mechanistic perspective, adherence to MMF may alleviate caregiving stress and enhance mothers’ confidence in their child’s nutritional status, thereby helping stabilize key hormonal pathways-particularly melatonin-, oxytocin-, and cortisol-related regulation-that support circadian alignment and sleep quality.^20,24,49 In contrast, multiparous mothers, who carry greater cumulative caregiving demands, may find it more difficult to maintain MMF, a strain that can heighten stress and dysregulate cortisol secretion through the HPA axis, increasing vulnerability to anxiety.⁵¹ Studies show that half of mothers reported adding cereal to bottles,⁴⁷ and fewer than 5% mentioned feeding eggs, meat, legumes, or nuts. In Chongqing, cereals were the most common first food (77.3%).³⁹ Additionally, we observed an interaction between child age and feeding: among mothers of 12–23-month-olds, meeting MMF and ZVF standards significantly reduced the risk of sleep disturbances compared to mothers of younger infants. The observed reduction in sleep disturbances could stem from decreased infant gastrointestinal distress (due to restricted fermentable fibres) and enhanced calming effects of frequent feedings, collectively improving infant-parent sleep continuity.^52,53 Moreover, increased infant age is generally considered a protective factor against maternal sleep disruption⁵⁴ and maternal sleep self-regulation and improvement.⁵⁵

While breastfeeding has been extensively studied for its impact on maternal health, research on the effects of IYCF-CF practices on postpartum maternal mental health remains limited. This study highlights a novel finding: adequate IYCF practices may help alleviate the dual burden of anxiety and poor sleep commonly experienced by mothers. A key strength of this study lies in its methodological advancements over prior research. We not only employed validated sleep scales to improve the measurement and comparability of sleep quality but also incorporated maternal anxiety to explore the effects of complementary feeding on maternal health from multiple dimensions. Moreover, the use of nationally standardized IYCF indicators to assess feeding practices ensures a consistent and objective benchmark. Our study offers a conceptual framework that may help explain how complementary feeding behaviors relate to maternal sleep. By considering caregiving demands, stress-response processes, and circadian regulation as potential pathways, these findings provide a plausible lens through which to interpret the observed associations between feeding practices and sleep outcomes. This perspective also has implications for IYCF programming: aligning feeding schedules with developmentally appropriate meal frequency may help stabilize infant routines and, in turn, support more predictable maternal sleep opportunities. Incorporating sleep-focused guidance into IYCF counselling and maternal-child health services could therefore enhance program relevance by addressing both nutritional practices and their potential spillover effects on maternal well-being. However, several limitations should be considered when interpreting these findings. First, the cross-sectional design restricts causal inference and limits understanding of the temporal dynamics linking complementary feeding behaviors with maternal sleep quality and anxiety. Longitudinal studies, ideally incorporating mechanistic approaches to clarify biological and behavioral pathways, are needed to establish directionality and capture changes across the complementary feeding period. Second, although validated instruments were employed, reliance on self-reported feeding practices and subjective sleep measures may introduce information bias. Future research should integrate objective sleep-monitoring tools (eg, actigraphy) and digital dietary tracking systems to enhance measurement accuracy. Third, 90.1% of participating mothers held a college degree or higher, representing a highly educated subgroup within the specific context of urban China. This limits the generalizability of findings to populations with different educational, socioeconomic, or cultural backgrounds-particularly in settings where infant‑feeding practices and maternal roles may be shaped by distinct cultural norms and family structures. Broader recruitment across more diverse maternal populations will be essential to strengthen external validity and advance understanding of behavioral mechanisms linking infant caregiving and maternal sleep health.

Conclusion

In this study from Chongqing, a major metropolis in Southwest China, we examined highly educated mothers of children aged 6–23 months. Despite this profile, maternal sleep disturbances (27.1%) and anxiety (~20%) remained common, and adherence to core IYCF practices-including MDD, MMF, MAD, and EFF-was lower than in comparable studies. Only 12.5% of children met MAD intake recommendations.

Improved adherence to MMF was linked with lower maternal anxiety and fewer sleep disturbances. Although causal inference is not possible in this cross-sectional design, these associations may reflect underlying biobehavioral pathways. More regular feeding routines may improve maternal sleep through multiple pathways. In addition to supporting hormonal regulation-particularly cortisol, melatonin, and oxytocin, which are involved in circadian alignment and stress buffering-adequate feeding practices may also promote infant satiety, reduce gastrointestinal discomfort, and enhance infant sleep continuity. These effects may be especially pronounced among older infants, for whom improved feeding regularity and maturation-related sleep consolidation jointly reduce nocturnal caregiving demands and support maternal sleep quality. Embedding brief maternal sleep support into IYCF counselling and postpartum services, following the global complementary food addition recommendation guidelines, may better address the interconnected nutritional and sleep needs of mothers and infants. Such integration could enhance caregiver well-being while strengthening the effectiveness of complementary feeding programs.