Association between mealtime regularity and breakfast frequency with postpartum depression in first-time Japanese mothers: a cross-sectional study

Abstract

Background

Postpartum depression (PPD) is a common mental health condition that negatively affects maternal wellbeing and child development. Although daily lifestyle behaviors have been implicated in PPD, the roles of breakfast skipping and mealtime regularity, which are key components of chrononutrition, remain insufficiently understood. This study examined the associations between mealtime regularity and postpartum depressive symptoms in first-time Japanese mothers.

Methods

The cross-sectional study was conducted through an online survey and included 841first-time Japanese mothers with a child aged 0–12 months. Depressive symptoms were assessed using the Edinburgh Postnatal Depression Scale (EPDS). Mealtime regularity and breakfast frequency were evaluated using self-administered questionnaires. Furthermore, associations between EPDS scores and eating-related behaviors were examined using generalized estimating equations, after adjusting for sociodemographic factors, health-related behaviors, mothers’ and infants’ sleep characteristics, and partners’ working days.

Results

Mothers with regular mealtime had significantly lower EPDS scores than those with irregular mealtime. After full adjustment for covariates, regular mealtime remained independently associated with lower EPDS scores, whereas breakfast skipping was not.

Conclusion

Regular mealtime, but not breakfast consumption, was independently associated with fewer postpartum depressive symptoms. These findings suggest that the temporal organization of eating behaviors may represent a novel and modifiable lifestyle factor relevant to postpartum mental health. Further longitudinal and interventional studies are warranted to clarify causality and evaluate mealtime regularity as a potential target for PPD prevention.

Trial registration

UMINID: 000051573; date of registration: July 10, 2023.

Background

Postpartum depression (PPD) is a common mental health disorder that affects approximately 17% of mothers worldwide [1]. PPD has serious consequences on maternal well-being, mother–infant bonding, and child development [2, 3]. PPD is also highly prevalent in Japan, affecting approximately 10–15% of postpartum mothers [4, 5]. PPD has been variably defined as a condition occurring between 4 weeks and 12 months after childbirth [6]. However, depressive symptoms frequently emerge during pregnancy rather than after childbirth, underscoring the importance of identifying modifiable lifestyle-related risk factors associated with PPD initiation and severity [7]. While psychological therapies and antidepressants effectively treat PPD, access to these treatments remains limited, and preventive strategies targeting lifestyle behaviors are still inadequate [8].

PPD is associated with daily lifestyle behaviors, such as sleep, physical activity, and food and nutrition. Sleep disturbance, including fragmented sleep, poor sleep quality, and frequent nighttime awakenings related to infant care, has been consistently reported as a risk factor for PPD [9–11]. Meanwhile, regular physical activity and exercise have been associated with a lower risk and reduced severity of PPD symptoms [12].

Dietary habits have also been implicated in postpartum mental health issues. A systematic review found that greater adherence to a healthy postpartum diet was generally associated with lower PPD symptoms [13, 14]. In Japanese populations, healthier dietary habits and overall healthier dietary patterns are also associated with lower depressive symptoms during pregnancy and a reduced risk of PPD [15, 16].

Concurrently, chrononutrition, which focuses on the time and regularity of food intake in relation to circadian rhythms, has emerged as an important determinant of mental and metabolic health [17]. Although breakfast skipping has been linked to depression and mental health outcomes in the non-pregnant population, evidence regarding its association with PPD remains limited [18–20]. A recent cross-sectional study of pregnant women in Korea found that skipping breakfast was significantly associated with higher PPD scores [21]. Additionally, associations between irregular mealtime and poor mental health have been reported in the non-pregnancy population [22, 23]. A cross-sectional study of Japanese women four months postpartum found that irregular eating patterns, such as “not eating meals regularly,” were significantly associated with poorer mental health status [24]. Furthermore, women with poor sleep, depression, anxiety, and stress showed significantly higher odds of unhealthy eating behaviors, including meal skipping and meal delaying [25]. Overall, the current evidence suggests potential associations between lifestyle behaviors and postpartum mental health. However, more studies are required to better understand the associations between modifiable daily behaviors and postpartum depression.

Therefore, this cross-sectional study investigated the association between mealtime regularity and PPD symptoms in first-time Japanese mothers of infants aged 0–12 months, after adjusting for sleep quality, health-related behaviors, infant sleep characteristics, and sociodemographic factors.

Methods

Ethical approval and data collection

This cross-sectional study explored the relationships among the postpartum mother’s mental health, social and environmental factors, and daily behaviors. Further, all methods used in this cross-sectional study complied with the STROBE statement [26]. This study also adhered with the guidelines of the Declaration of Helsinki and was approved by Hiroshima University’s Ethical Committee for Epidemiology on July 10, 2023 (No. E2023-0047; UMINID: 000051573). Subsequently, informed consent was provided by all participants in the surveys; they consented to the collection and use of their data for research. Aresearch company (Macromill Inc., Tokyo, Japan) was commissioned to conduct an online survey from Aug 1–4, 2023. The participants in the research company’s cohort lived in Japan. Finally, the company was instructed to collect samples with a 1:1 male-to-female ratio and ensure an even distribution across the ages (1–12 months old) of the babies.

Participants

A priori power analysis for multiple linear regression was conducted to determine the sample size. Assuming a two-sided α of 0.05, 80% power, and up to 10 covariates, a sample size of approximately 800 participants was sufficient to detect a small effect size (0.02). Although previous studies have reported no clear association between parity and postpartum depression [27, 28], multiparous women may have different physiological, behavioral, and psychosocial adaptations due to previous pregnancies and childcare experience, which could confound the associations of interest. To maintain homogeneity within a relatively small sample size, we restricted the study population to a more defined group. The inclusion criteria for the participants were as follows: (1) mother who had only a firstborn baby aged 0–12 months; (2) mother who was working before pregnancy; (3) mother who was not working when she answered this questionnaire; and (4) mother whose partner had exhausted childcare leave. The exclusion criteria comprised single mothers living with their baby. Our survey covered 1,030 participants. After excluding those with missing response variables (n = 18) and those who did not meet the inclusion criteria (mothers who were not pregnant: n = 67, mothers who were not working before pregnancy: n = 78; fathers who were not working: n = 7; mothers who were living with their baby without other family: n = 19), we analyzed a total of 841 participants.

Questionnaire construction

The sociodemographic variables included mother’s age, baby’s age (in months) and sex, and partner’s working days (days/weeks). Data were also obtained on whether the pregnancy resulted from infertility treatment, the duration of pregnancy, and presence or absence of cesarean delivery. The mothers’ anthropometric data, including weight and height, were collected to calculate Body Mass Index (BMI). Their personality data were collected using the Big-Five personality traits [29]. Each personality trait (extraversion, agreeableness, conscientiousness, neuroticism, and openness) was calculated as the average of the scores on two related questions (average score range: 1–7).

The Japanese version of the Edinburgh Postnatal Depression Scale (EPDS) was used to assess postpartum depression [30]. The EPDS score was calculated as the sum of the 10 questions included in the scale. An EPDS score of ≥ 9 was used as an indicator of the occurrence of PPD symptoms. The Oguri-Shirakawa-Azumi Sleep Inventory Middle-aged and Aged version (OSA-MA) was used to evaluate sleep problems, which were defined by participants’ lower scores [31, 32]. Behaviors related to health markers were assessed based on the number of days per week of breakfast intake, nighttime snack intake, alcohol consumption, caffeinated beverage intake, smoking, and exercise. Mealtime regularity was assessed using the question, “Do you currently eat meals at the same time each day?” with a five-point answer scale (1 = strongly disagree; 2 = disagree; 3 = neither agree or disagree; 4 = agree; and 5 = strongly agree) [22]. Baby’s awakenings during nighttime sleep (times/day) and sleep latency with the mother (in minutes) were also examined as risk factors for PPD.

Statistical analyses

Data were analyzed using IBM SPSS Statistics (version 29.0; IBM Ltd., Armonk, NY, United States). Descriptive statistics are presented as means and standard deviations. During this study, we divided two groups based on mealtime regularity (scores of 1–3 for “irregular or neither” and 4–5 for “regular”), breakfast intake (0–6 days/week or 7 days/week), and EPDS score (score of less than 9 for without symptom and ≥ 9 for PPD). The Mann–Whitney U test was used to examine significant differences between the groups. As EPDS scores did not show a normal distribution, generalized estimating equations (GEE) were selected to test the association between EPDS scores (continuous variable) and mealtime regularity (0 = irregular or neither; 1 = regular) and/or breakfast frequency (0 = 0–6 days/week; 1 = 7 days/week). For the GEE, covariates (mother’s age, her BMI, personality, OSA-MA, frequency of nighttime snacking, alcohol consumption, caffeinated beverage intake, smoking, and exercise, baby’s age, baby’s nighttime awakenings and sleep latency, and partner’s working days) were adjusted. In Model 1, mealtime regularity and breakfast frequency were analyzed separately, whereas in Model 2, both variables were included simultaneously. Quasi-likelihood under the independence model criterion (QIC) was used to evaluate the relative goodness-of-fit among each model, and to support model comparison. Statistical significance was set at p < 0.05.

Results

Participants’ characteristics based on mealtime regularity and breakfast frequency

Participant characteristics based on mealtime regularity, breakfast frequency, and EPDS scores are summarized in Table 1. 400 participants had irregular or neither mealtime regularity and 441 participants had regular mealtimes. For breakfast intake, 318 participants reported the frequency of 0–6 days/week and 523 participants ate breakfast every day. Based on EPDS scores, 523 participants were classified as lower (< 9) and 318 as higher (≥ 9). Mothers with regular mealtime were slightly older than those with irregular or neither mealtime (31.1 ± 4.2 vs. 30.4 ± 4.2 years, p = 0.02) and had significantly lower EPDS scores (6.9 ± 5.7 vs. 8.6 ± 5.7, p < 0.001). Regular mealtime was also associated with more favorable sleep-related outcomes, including higher scores for sleepiness on rising, initiation and maintenance of sleep, refreshing on rising, and sleep length, as assessed using the OSA-MA sleep inventory (all p < 0.05). Mothers who reported daily breakfast consumption (7 days/week) were older, had lower BMI, and exhibited lower EPDS scores compared with those who consumed breakfast 0–6 days/week (7.3 ± 5.6 vs. 8.3 ± 6.0, p = 0.021). Significant differences were also observed in several lifestyle behaviors, including alcohol consumption, smoking, and exercise frequency (all p < 0.01). No significant differences were found between the groups regarding delivery outcomes or infant characteristics, except for baby’s nighttime awakenings (p = 0.039). Further, breakfast intake frequency was significantly higher in the regular mealtime group, compared with the irregular mealtime group (6.1 ± 1.9 vs. 4.9 ± 2.5 days/week, p < 0.001). These results indicated that mealtime regularity and breakfast consumption were closely related behaviors.

Table 1.

Characteristics of participants by mealtime regularity, breakfast frequency, and EPDS scores

				Mealtime regularity					Breakfast intake					EPDS scores
		Total		Irregular or neither		Regular			0–6 days/week		7 days/week			Lower (< 9)		Higher (≥ 9)
		n = 841		n = 400		n = 441			n = 318		n = 523			n = 523		n = 318
		n (%)		n (%)		n (%)		p	n (%)		n (%)		p	n (%)		n (%)		p
Pregnancy by fertilization	No	647 (76.9)		312 (78.0)		335 (76.0)		0.512	257 (80.8)		390 (74.6)		0.043	401 (76.7)		246 (77.4)		0.866
	Yes	194 (23.1)		88 (22.0)		106 (24.0)			61 (19.2)		133 (25.4)			122 (23.3)		72 (22.6)
Gestation period	Full-term delivery	778 (92.5)		376 (94.0)		402 (91.1)		0.149	297 (93.3)		481 (91.9)		0.501	496 (94.8)		282 (88.7)		< 0.001
	Preterm or postterm delivery	63 (7.5)		24 (6.0)		39 (8.8)			21 (6.6)		42 (8.0)			27 (5.2)		36 (11.3)
Cesarean section	No	685 (81.5)		328 (82.0)		357 (80.9)		0.723	253 (79.5)		432 (82.6)		0.274	425 (81.3)		260 (81.8)		0.466
	Yes	156 (18.5)		72 (18.0)		84 (19.0)			65 (20.4)		91 (17.3)			98 (18.7)		58 (18.2)

		Mean	SD	Mean	SD	Mean	SD	p	Mean	SD	Mean	SD	p	Mean	SD	Mean	SD	p
Mother’s age (years)		30.8	4.2	30.4	4.2	31.1	4.2	0.02	29.8	4.3	31.4	4.1	< 0.001	30.8	4.1	30.9	4.5	0.654
Baby’s age (months)		6.4	3.5	6.1	3.4	6.7	3.6	0.006	6.1	3.4	6.6	3.6	0.063	6.1	3.5	6.8	3.6	0.005
BMI (kg/㎡)		21.0	3.2	21.1	3.4	21.0	3.0	0.715	21.5	3.4	20.7	3.0	< 0.001	21.0	2.9	21.1	3.5	0.653
Personality_Extraversion (score)		4.1	1.4	4.1	1.4	4.1	1.4	0.905	4.1	1.4	4.1	1.4	0.766	4.2	1.4	3.8	1.3	< 0.001
Personality_Agreeableness (score)		4.9	1.0	4.8	1.1	5.0	1.0	0.001	4.8	1.0	5.0	1.0	0.003	5.1	1.0	4.7	1.0	< 0.001
Personality_Conscientiousness (score)		3.8	1.2	3.7	1.2	3.8	1.2	0.14	3.6	1.1	3.9	1.2	< 0.001	3.8	1.2	3.8	1.1	0.498
Personality_Neuroticism (score)		4.3	1.2	4.5	1.2	4.2	1.2	0.007	4.4	1.2	4.3	1.2	0.507	4.1	1.2	4.7	1.1	< 0.001
Personality_Openness (score)		3.6	1.1	3.6	1.1	3.6	1.2	0.863	3.7	1.1	3.6	1.1	0.068	3.6	1.2	3.6	1.1	0.99
EPDS scores		7.7	5.8	8.6	5.7	6.9	5.7	< 0.001	8.3	6.0	7.3	5.6	0.021	-	-	-	-	-
OSA-MA Sleepiness on rising (Zc score)		15.4	5.9	14.7	5.9	16.1	5.9	0.001	15.1	5.8	15.6	6.0	0.134	16.9	5.6	13.1	5.6	< 0.001
OSA-MA Initiation and maintenance of sleep (Zc score)		13.2	5.8	12.7	5.6	13.7	6.0	0.019	13.0	5.8	13.3	5.8	0.984	14.4	6.1	11.3	4.9	< 0.001
OSA-MA Frequent dreaming (Zc score)		16.2	8.6	16.1	8.4	16.3	8.8	0.728	15.8	8.4	16.5	8.7	0.325	17.3	8.6	14.5	8.4	< 0.001
OSA-MA Refreshing on rising (Zc score)		12.6	6.3	11.7	6.0	13.4	6.4	< 0.001	11.9	6.5	12.9	6.2	0.024	14.1	6.3	10.0	5.5	< 0.001
OSA-MA Sleep length (Zc score)		19.4	6.8	18.6	6.9	20.2	6.7	< 0.001	18.8	7.3	19.7	6.5	0.06	20.7	6.5	17.2	6.7	< 0.001
Mealtime regularity (score)		3.2	1.2	-	-	-	-	-	2.7	1.2	3.5	1.1	< 0.001	3.3	1.2	3.0	1.2	0.002
Breakfast intake (days/week)		5.5	2.3	4.9	2.5	6.1	1.9	< 0.001	-	-	-	-	-	5.6	2.1	5.3	2.4	0.042
Nighttime snack intake (days/week)		2.4	2.7	2.4	2.7	2.3	2.7	0.294	2.3	2.5	2.4	2.8	0.362	2.4	2.7	2.4	2.7	0.991
Alcohol intake (days/week)		0.5	1.3	0.5	1.4	0.5	1.3	0.915	0.8	1.7	0.3	1.0	< 0.001	0.5	1.3	0.6	1.4	0.333
Caffeine intake (days/week)		2.8	2.7	2.8	2.7	2.8	2.7	0.944	2.8	2.5	2.8	2.8	0.283	2.9	2.7	2.6	2.7	0.197
Smoking (days/week)		0.3	1.3	0.4	1.5	0.2	1.0	0.155	0.6	1.8	0.1	0.7	< 0.001	0.1	0.9	0.5	1.7	< 0.001
Exercise (days/week)		0.8	1.6	0.7	1.5	0.8	1.6	0.914	0.9	1.8	0.7	1.5	0.008	0.6	1.5	1.0	1.7	0.006
Baby’s awakenings during nighttime sleep (times/day)		1.5	1.2	1.6	1.2	1.4	1.2	0.105	1.4	1.2	1.6	1.2	0.039	1.4	1.2	1.6	1.2	0.023
Duration of baby’s sleep latency with mother (min)		32.9	21.6	35.9	22.7	30.2	20.1	< 0.001	34.8	24.5	31.8	19.5	0.222	31.3	20.3	35.7	23.3	0.004
Partner’s working days (days/week)		5.1	0.8	5.2	0.8	5.1	0.8	0.011	5.1	1.0	5.1	0.6	0.231	5.2	0.7	5.0	0.9	0.017

The Mann–Whitney U test was used to examine significant differences between the groups

Significant data (P < 0.05) are shown in bold

Mothers with higher EPDS scores showed a higher frequency of preterm or postterm deliveries. Additionally, they tended to have older infants, and displayed certain personality traits (lower extraversion and agreeableness, and higher neuroticism). These mothers had lower OSA-MA sleep scores across all domains, irregular mealtime, lower breakfast frequency, higher smoking rates, and higher exercise frequency. Consequently, their babies experienced higher nighttime awakenings, longer sleep onset latency. Finally, the mother’s partner tended to have fewer working days (Table 1).

Associations between mealtime regularity, breakfast frequency, and EPDS scores

Associations among mealtime regularity, breakfast frequency, and EPDS scores were examined using GEE (Table 2). In Model 1, mother’s age, her BMI, personality, pregnancy conditions, and health-related behaviors (sleep, nighttime snacking, alcohol consumption, caffeinated beverage intake, and smoking) were used as covariates. Covariates also covered baby’s age, sex, and sleep condition, and the partner’s working days. Regular mealtime was significantly associated with lower EPDS scores (B = − 0.71; 95% CI − 1.37 to − 0.04; p = 0.036), whereas breakfast frequency was not significantly associated with EPDS scores (B = 0.03; 95% CI − 0.68 to 0.75; p = 0.925). In Model 2, by analyzing both mealtime regularity and breakfast frequency in the same equation, regular mealtime remained significantly associated with lower EPDS scores (B = − 0.76; 95% CI − 1.45 to − 0.07; p = 0.031). By contrast, breakfast frequency remained non-significant (B = 0.23; 95% CI − 0.52 to 0.98; p = 0.545). The QIC indicated a comparable fit across the models. Covariates, including baby’s sex and age, gestation period, mother’s personality (extraversion, agreeableness, and neuroticism), OSA-MA scores, smoking and exercise frequency, and partner’s working days were also significantly associated with EPDS scores in Model 2.

Table 2.

Association between mealtime regularity and/or breakfast frequency with EPDS scores using generalized estimating equation models

				95% CI
	QIC	Independent variables	B	Lower	Upper	Wald Chi-Square	p
Model 1	17436.3	Mealtime regularity	-0.71	-1.37	-0.04	4.402	0.036
	17530.6	Breakfast frequency	0.03	-0.68	0.75	0.009	0.925
Model 2	17430.1	Mealtime regularity	-0.76	-1.45	-0.07	4.66	0.031
		Breakfast frequency	0.23	-0.52	0.98	0.367	0.545

Mealtime regularity was divided into irregular/neither and regular categories. Breakfast frequency was divided into 0–6 days/week and 7 days/week. Irregular/neither mealtime or 0–6 days/week breakfast intake were used as a reference in the generalized estimating equation (GEE) models. Adjustment included age, BMI, baby’s age (months), baby’s sex, pregnancy by fertilization, gestation period, cesarean section, Big Five personality traits, OSA score, nighttime snack intake, alcohol intake, caffeine intake, smoking, baby’s awakenings during night-time sleep, duration of baby’s sleep latency with mother, and partner’s working days. In Model 1, mealtime regularity and breakfast frequency were analyzed separately, whereas in Model 2, both variables were included simultaneously

QIC quasi-likelihood under the independence model criterion, B unstandardized regression coefficient, CI confidence interval

Significant data (P < 0.05) are shown in bold

Discussion

In this cross-sectional study of first-time Japanese postpartum mothers, regular mealtime was independently associated with lower EPDS scores, whereas breakfast frequency was not. This association remained significant after adjusting for sociodemographic factors, health-related behaviors, and mother and infant sleep characteristics. A previous study has reported an association between regular eating habits and postpartum mental health; however, the study did not clearly define what constituted “regular eating” and did not adjust for breakfast skipping [24]. To our knowledge, this study is the first to demonstrate an association between irregularity in meal timing and PPD symptoms.

In this study, no significant association was observed between breakfast skipping and PPD symptoms. Although numerous studies have reported an association between breakfast skipping and mental health in the general population, evidence regarding the relationship between breakfast skipping and perinatal mental health remains limited [19, 21]. Breakfast skipping during or after pregnancy has been associated with inadequate nutritional intake, gestational diabetes, and adverse neurodevelopmental outcomes in the offspring [33–36]. Furthermore, the inadequate intake of key nutrients has been linked to PPD symptoms, underscoring the importance of regular breakfast consumption during the perinatal period [13–16]. Conversely, habitual breakfast skipping may result in regular daily fasting intervals, which could potentially exert beneficial effects on physical and mental health. Several intervention studies in non-pregnant populations have demonstrated that intermittent fasting and time-restricted eating are associated with improved mental health outcomes [18, 37]. Intermittent fasting has also been suggested to be effective in improving mental health, especially in relation to gestational diabetes [38]. Together, these factors may attenuate or obscure the association between breakfast consumption and mental health during the postpartum period in the present study.

Irregular eating patterns have recently emerged as an important lifestyle factor associated with mental health and metabolic outcomes. Irregular mealtime is linked to poor mental health, including higher levels of psychological distress and anxiety, and stressful life events [39–41]. Irregular mealtime has been associated with poor sleep quality and increased daytime sleepiness, suggesting a close relationship between disrupted eating rhythms and sleep disturbances [42]. Infant age may influence maternal behavioral rhythms, including sleep and mealtime regularity. In fact, our data showed the significant relationship between mealtime regularity and baby’s age (Table 1). Previous studies have shown that maternal sleep variability is closely linked to infant sleep patterns and changes as infants mature, with greater irregularity in the early postpartum period and increased stabilization over time [43]. In this study, irregular mealtime was directly associated with PPD symptoms, even after adjusting for a wide range of potential confounders (including maternal sleep disturbance and baby’s age). Beyond mental health and sleep, irregular mealtime has been associated with obesity and cardiovascular risk among shift and non-shift workers [44–46]. Moreover, discrepancies in mealtime between weekdays and weekends—referred to as “eating jet lag”—have been linked to BMI and lifestyle-related diseases independent of sleep irregularity [47–49]. One plausible mechanism by which irregular mealtime influences mental health is disruption of the circadian clock. The circadian clock is a fundamental system that maintains physiological homeostasis and is entrained by daily environmental cues, particularly light exposure and food intake [50]. Irregular mealtime provides inconsistent zeitgebers (entrainment stimulation) to the circadian system, leading to circadian misalignment [51]. Circadian disruption is associated with an increased risk of various health conditions, including metabolic disorders and other lifestyle-related diseases [52]. Collectively, these findings highlighted irregular mealtime as a novel and potentially modifiable behavioral factor relevant to mental, sleep, and cardiometabolic health.

This study had several limitations. Owing to the cross-sectional design, causal relationships between irregular mealtime and postpartum depressive symptoms could not be established. The use of an online self-administered survey may have introduced selection and reporting bias. In addition, the proportion of participants with an EPDS score ≥ 9 in this study (37.8%) was higher than that reported in the general population, suggesting potential selection bias. The study population was limited to first-time Japanese mothers with prior work experience who were not currently employed, which may limit the generalizability of the findings. Although household income is known to be associated with postpartum depression, it was not included as a covariate due to a high proportion of missing data (26.6%). Furthermore, irregular mealtime was assessed using a simplified measure. Future studies should incorporate more detailed assessments of mealtime variability and examine irregular meal skipping and irregular nutrient intake in relation to postpartum depressive symptoms.

Conclusions

In conclusion, this cross-sectional study demonstrated that irregular mealtime was independently associated with PPD symptoms among first-time Japanese mothers, even after accounting for sleep disturbances and other relevant confounding factors. Further prospective and interventional studies are needed to clarify the causal relationship and determine whether interventions targeting mealtime regularity may contribute to the prevention or management of postpartum depression.

Abbreviations

PPD

Postpartum depression

EPDS

Edinburgh Postnatal Depression Scale

BMI

Body Mass Index

OSA-MA

Oguri-Shirakawa-Azumi Sleep Inventory Middle-aged and Aged version

GEE

Generalized estimating equations

QIC

Quasi-likelihood under the independence model criterion

B

Unstandardized regression coefficient

CI

Confidence interval

Authors’ contributions

Research conception and design: YT, YM, SS; methodology and data collection: YT, YM, SS; statistical analysis of the data: YT, YM, YL; writing of the manuscript: YT; supervision: TK, SS.

Funding

This work was partially supported by the Combi Corporation and JST-FOREST Program (JPMJFR205G for Y.T.).

Data availability

These data will be provided to the researchers upon request for research purposes.

Declarations

Ethics approval and consent to participate

This study adhered with the guidelines of the Declaration of Helsinki and was approved by Hiroshima University’s Ethical Committee for Epidemiology on July 10, 2023 (No. E2023-0047; UMINID: 000051573). Subsequently, informed consent was provided by all participants in the surveys; they consented to the collection and use of their data for research.

Consent for publication

Not applicable.

Competing interests

Y.M. is a full-time employee of the Combi Corporation, which conducted the online survey and provided financial support for this study to the other authors.