Sleep During Pregnancy and Offspring Outcomes From Infancy to Childhood: A Systematic Review

Abstract

Objective:

Prenatal sleep problems have been previously linked to poor birth outcomes. However, much less is known about the associations between poor prenatal maternal sleep and offspring outcomes after birth. The purpose of this systematic review was to synthesize the findings linking prenatal maternal sleep health and offspring health and development during infancy and childhood.

Methods:

A total of 4650 nonduplicate articles were identified via PsycInfo and PubMed databases. After screening and full-text review, 34 articles met the inclusion criteria and were extracted for information.

Results:

The bulk of studies in this review (n = 19; 76.5%) were published between 2019 and 2024. Measures of prenatal sleep included sleep timing, quality, sleep disorders and/or symptoms of disorders, and daytime sleepiness. Offspring outcomes were categorized as follows: a) sleep health (e.g., nighttime sleep duration, night wakings), b) physical health (e.g., body mass index, hospitalizations), c) child developmental outcomes (e.g., global development, negative affect, executive functioning), and d) brain structure and function (e.g., brain volume, event-related potentials). Evidence consistently links poor prenatal sleep health to poorer offspring sleep, higher body mass index, higher prevalence of physical health conditions, poorer global development, and more behavioral problems. Emerging evidence also links prenatal sleep to differences in offspring brain structure and function.

Conclusions:

Poor prenatal maternal sleep health may be an environmental signal that informs offspring health. Future studies are needed to fully understand the pervasive, intergenerational, and long-lasting effects of sleep across pregnancy.

INTRODUCTION

Sleep problems are a pressing public health concern. Twenty-one percent of Americans report a chronic sleep disorder (1), and sleep problems contribute to a myriad of physical and psychological diseases (2–4). Notably, women experience poorer sleep than men across the lifespan (5–7) and are at greater risk of developing depression and cardiovascular diseases following sleep problems (8–10). Despite established gender disparities in sleep health (11–13), women’s sleep is understudied (14–16). This gap may be particularly problematic during pregnancy when rates of sleep problems are high (17) and have intergenerational consequences (18–20).

Sleep during pregnancy plays a critical role in regulating maternal endocrine and metabolic systems that promote homeostasis (21,22). Further, poor sleep during pregnancy is associated with greater risk of prenatal depression and mood disorders (23,24), gestational diabetes and preeclampsia (25,26), and prenatal inflammation (27,28). Prenatal maternal sleep not only affects maternal health but also regulates fetal developmental processes, including growth, parturition, and entrainment of circadian rhythms (29–31). In sum, prenatal maternal sleep plays a key role in both maternal and fetal health.

The Developmental Origins of Health and Disease (DOHaD) Hypothesis posits that one’s health and development are especially susceptible to environmental signals during periods of rapid growth (i.e., sensitive periods; ((32,33))). In line with DOHaD, it is plausible that poor prenatal maternal sleep sculpts offspring development. Existing evidence supports prenatal sleep disturbances as a salient prenatal signal that lead to poor birth outcomes, including shortened gestation (34), preterm birth (35), low birth weight, and decreased birth length (36). Consequently, previous systematic and scoping reviews have examined prenatal sleep and birth outcomes (e.g., preterm birth and low birth weight), but have not considered developmental consequences beyond birth (18,37). Poor birth outcomes are associated with negative health and developmental outcomes over the lifespan. However, few empirical studies directly test the consequences of prenatal sleep on developmental outcomes beyond birth. The goal of this review is to synthesize the literature investigating the links between prenatal maternal sleep problems and offspring developmental outcomes from infancy through childhood and adolescence to inform future directions for prenatal sleep research and improve the evidence base for prenatal sleep interventions. This is the first systematic review, to our knowledge, to synthesize the literature linking prenatal maternal sleep and offspring outcomes after birth.

The Current Review

The first aim was to determine if prenatal sleep is associated with offspring outcomes from infancy to adolescence. The second aim was to identify which offspring outcomes have been most frequently studied in relation to prenatal sleep. Finally, as sleep is a multidimensional construct, the third aim was to identify which indicators of prenatal sleep (e.g., timing, quality) are most consistently associated with offspring outcomes. Because no consistent guidance exists on a scientific definition of adequate sleep (38), we define poor sleep based on the findings from the present review. Poor sleep encompasses short sleep duration, late bedtime and chronotype, decreased sleep quality or increased sleep problems/disturbances, sleep disorders or symptoms of sleep disorders, and daytime sleepiness.

METHODS

Using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (39), we conducted a systematic search in PubMed and PsychINFO (EBSCOHost) databases, from the earliest record available to January 2024. We also conducted an additional step of reviewing references of included articles to identify any new articles. This systematic review was not preregistered.

Search terms were identified based on prior systematic reviews on prenatal sleep and birth outcomes (18,37) and studies related to pregnancy, sleep health, and offspring outcomes (e.g., (28,33)). Search terms were finalized using a deductive process with a university librarian in which approximately 30 iterations of terms were tested (Supplemental Digital Content, Table S1, http://links.lww.com/PSYMED/B50). The final string of search terms yielded the largest number of publications out of all attempted iterations. As a result, the following final search terms were utilized: (prenatal sleep OR perinatal sleep OR antenatal sleep OR sleep pregnan* OR gestation* sleep) AND (child* OR adolesc* OR infan* OR youth OR pediatr*). The human population and English language filters were applied within each search database.

Eligibility Criteria

Inclusion and exclusion criteria were determined prior to screening studies. Studies included in this review were a) published peer-reviewed articles written in English, b) empirical studies that analyzed original data with human research participants, c) findings that examined sleep during pregnancy at least once (prospectively or retrospectively), and d) those that included at least one outcome assessed during infancy through 18 years of age. Studies were excluded if they were not peer-reviewed or empirical studies (e.g., book chapters, review, dissertations) or included only birth outcomes (e.g., gestational length, birth weight).

Study Selection

From our search terms, 4886 records were identified via databases, 8 records were identified by reviewing references of retrieved studies, and 1 study was identified via Google Scholar alerts (Figure 1). Studies were imported into Covidence systematic review software for screening, and after removing 245 duplicates, 4650 articles remained. Four coders (M.N.B., D.H., E.T., and P.K.) independently reviewed the first 100 articles and discussed discrepancies until consensus was achieved. Then, coders screened titles and abstracts of the articles in pairs (96.2% interrater reliability across all four coders). Disagreements were discussed and resolved among coders, consulting with senior authors (E.P.D. and J.R.D.) as needed. Abstract and title screening revealed that 4570 articles were not relevant for this review (see PRISMA Diagram, Figure 1). The main reason for exclusion at this stage was absence of assessment of sleep in the pregnant individual (e.g., only measured infant sleep). The 80 remaining articles were independently inspected at the full-text level by the first and second authors (M.N.B. and D.H.; 86.3% interrater reliability). Disagreements were reconciled between the two coders, consulting senior authors as needed. A total of 46 articles were excluded for the following reasons: did not assess prenatal sleep (n = 17), did not measure child outcomes (n = 15), did not examine the association between prenatal sleep and offspring outcomes (i.e., the paper included both variables but not their association; n = 8), was not an empirical study (e.g., literature review, chapter, or protocol; n = 4), or were dissertations (not in a peer-reviewed journal; n = 2). The remaining 34 articles meeting the review criteria were included in this review.

FIGURE 1.

PRISMA diagram.

Data Extraction

Using Covidence software, we developed a data extraction form to identify study characteristics and findings. Our data extraction assessed sample characteristics, study design, how prenatal sleep was measured, type of child outcome, and study results (see Supplemental Digital Content, Table S3, for a detailed list of variables that were coded, http://links.lww.com/PSYMED/B50). The data extraction form was piloted using five randomly selected studies and refined accordingly. Two reviewers (M.N.B. and D.H.) independently extracted data for all included studies, and conflicts were resolved through discussion. Two additional reviewers (E.T. and P.K.) double-checked all components of data extraction while creating detailed tables with study information.

All prenatal maternal sleep constructs were included in this study. Following full-text review, we categorized prenatal maternal sleep constructs as follows: a) sleep timing (e.g., sleep duration, bedtime, midpoint), b) sleep quality (e.g., disturbances), c) sleep disorders or symptoms of disorders (e.g., insomnia, apnea), and d) daytime sleepiness. In terms of outcomes, all offspring outcomes after birth were considered. We were also inclusive of all methods and timepoints for outcome measurement, including cases where the same outcome was measured multimodally (e.g., parent-report and objective) or at multiple timepoints. Upon completion of the full-text review and data extraction, we categorized child outcomes as follows: a) offspring sleep health, b) physical health, c) developmental outcomes, and d) brain structure and function. For studies that assessed multiple indicators of prenatal sleep (e.g., duration and quality) and/or child outcomes (e.g., child sleep and socioemotional problems), we reported on all available associations.

Included studies used a diverse array of statistical methods and covariates. Regarding covariates, we reported on findings from both unadjusted and adjusted models in our tables. However, because most study results did not differ between unadjusted and adjusted models, we only discussed findings from adjusted models if the results changed as a function of covariates (e.g., significant associations in unadjusted models did not hold after covariates or associations only became significant after covariates). There were two studies that did not report on direct associations between prenatal sleep and child outcomes but examined prenatal sleep as a mediator in a larger pathway (40,41). For these studies, we reported on bivariate correlations between prenatal sleep and offspring outcomes.

Risk of Bias

We evaluated study quality and risk of bias for all included studies using criteria that were informed by online tools (e.g., riskofbias.info), previous systematic reviews (42), and published guidelines (43)). We assessed study quality and risk of bias using the following dimensions: a) properties of prenatal maternal sleep measurement, b) properties of offspring outcome measurement, c) approach to addressing missing data, and d) inclusion of pertinent covariates (Supplemental Digital Content Table S2, http://links.lww.com/PSYMED/B50). For assessment of the properties of prenatal maternal sleep measures, “good” ratings were assigned to studies employing objective measures of prenatal sleep (e.g., actigraphy), or a previously validated self-report measure of prenatal sleep if collected more than once during pregnancy, “fair” ratings were assigned to studies utilizing previously validated self-report measures of prenatal sleep that only assessed prenatal maternal sleep once, and “poor” ratings were assigned to studies using invalid and unreliable instruments regardless of how many times it was collected during pregnancy. The properties of offspring outcomes were assessed in the same way that the properties of prenatal maternal sleep assessment were (i.e., objective measures of outcomes, validated self- or parent-report measures, number of times outcomes were assessed). We also screened for approach of addressing missing data (44). Studies were assigned “good” ratings if they reported and accounted for missing data using statistical analyses (e.g., multiple imputation, full-information maximum likelihood), “fair” ratings if they reported minimal missing data (<5%) but did not account for missing data in analyses, and “poor” ratings if they did not report missing data or utilized listwise deletion to address missingness. Additionally, we evaluated inclusion of covariates for each study because multiple studies highlight that sociodemographic factors, and physical and mental health can disrupt sleep health (45–47). We screened studies for inclusion of covariates in the following categories: a) sociodemographic risk, b) physical health, and c) mental health. Studies received “good” ratings if two or more of the covariate categories were represented, “fair” if one category was represented, and “poor” if none of the categories were represented or if the study did not include covariates. Coders (M.N.B. and D.H.) independently assessed each article on study quality and risk of bias, and disagreements were resolved together.

RESULTS

Descriptive Information

Study Characteristics

The earliest study in this review dated back to 1998, but most studies (n = 26; 76.5%) were published in 2019 and beyond, indicating substantial growth of research on this topic over the last 5 years. Studies were conducted globally, including in Eastern Asia (n = 11), Europe (n = 8), the United States (n = 9), Australia (n = 2), Southeastern Asia (n = 2), and the Middle East (n = 2).

Sample Characteristics

Sample sizes ranged from 58 to almost 3 million mother-offspring dyads (median = 795.5). Further, nine studies reported on the pregnant individuals’ race and/or ethnicity. Offspring outcomes were collected from birth to 11 years, spanning infancy (≤1 year; n = 21), early childhood (2–5 years; n = 19), and middle childhood (6–11 years; n = 5). Eleven studies (32.4%) examined outcomes during more than one developmental period (e.g., infancy and early childhood, early to middle childhood). Of note, no studies examined outcomes during adolescence (≥12 years).

Prenatal Sleep Measures

Sleep during pregnancy was predominantly measured via self-report questionnaires only (n = 25; 73.5%), with only seven studies using objective measures, and two studies using both self-report and objective measures. Of the studies that used self-report questionnaires, 14 studies used one or more validated measures including the Pittsburgh Sleep Quality Index (PSQI; (48); n = 6), which was the most utilized measure. In contrast, 16 studies constructed their own (not previously validated) measures of prenatal sleep health. Objective measures of prenatal maternal sleep included overnight laboratory visits using polysomnography or ambulatory sleep technology (n = 3), diagnoses of a sleep disorder by a health care provider (n = 2), and activity tracked by wearable devices (n = 2).

Prenatal Sleep Constructs

We categorized prenatal maternal sleep as follows: a) sleep quality, which consisted of prenatal sleep problems and/or ratings of overall quality (n = 12), sleep depth (light versus deep; n = 2), and feeling when waking up (bad versus good; n = 2); b) sleep timing, such as sleep duration (nighttime and total sleep throughout the day; n = 15), bedtime (n = 2), midpoint (middle timepoint between bedtime and wake time; n = 2), chronotype (natural individual preference to sleep at a certain time; n = 1), night-to-night variability in sleep duration (n = 1), and circadian misalignment (n = 1); c) diagnoses or symptoms of prenatal sleep disorders, including insomnia (n = 3), sleep apnea (n = 3), and sleep-disordered breathing (n = 3); and d) daytime sleepiness (n = 4).

Frequency and Timing of Prenatal Sleep Assessment

Most studies (n = 27; 79.4%) measured sleep prospectively (i.e., during pregnancy), whereas five studies measured sleep retrospectively (i.e., reporting on prenatal sleep after birth) and two studies obtained data on sleep disorders via hospital records. In terms of frequency, most studies (n = 25; 73.5%) measured sleep once during pregnancy (including retrospective reports of one timepoint), most often during mid-to-late gestation (after 20 gestational weeks; n = 17). Only one study focused solely on early gestation (before 20 gestational weeks), whereas other studies asked about pregnancy generally (n = 4) or pooled participants across different gestational periods (n = 3). Four studies asked about sleep during two timepoints, during early and mid-to-late gestation, and five studies asked about sleep three separate times, during early, middle, and late gestation. Of the nine studies (16.7%) that measured sleep at multiple timepoints, five examined whether associations with offspring outcomes differed by timing of exposure to poor prenatal sleep (49–53). Notably, only three studies examined whether changes in sleep across gestation predicted offspring outcomes (49,51,54).

Prenatal Sleep and Offspring Outcomes

Offspring outcomes were categorized as follows: a) sleep health (n = 16; Table 1A), b) physical health (n = 10; Table 1B), c) developmental outcomes (n = 13; Table 2A), and d) brain structure and function (n = 3; Table 2B). Findings are organized below, first by offspring outcome category and then by indicators of prenatal sleep.

TABLE 1.

Studies With Offspring Sleep and Physical Health Outcomes

(A) Studies With Offspring Sleep Health Outcomes
Study	Sample Size, Country, & Maternal Age	Prenatal Sleep Measure	Frequency & Timing of Prenatal Assessment	Offspring Outcome(s)	Age of Offspring Outcome(s)	Findings	Covariates
Section I: Prenatal sleep timing
Kuroda et al. (2021)	1133 mother-infant dyads in Japan Mage = 32.0	Self-reported sleep duration regularitya	Retrospective report in mid-to-late pregnancy	Maternal report of average infant sleep durationa	1 mo	Short prenatal maternal nighttime sleep duration was associated with infants sleeping more during the day than at night.
Lugo-Candelas et al. (2023) b	794 mother-child dyads in the United States Mage = 32.3	Self-reported sleep duration using MSHPa and/or PSQI	3 times, once in each trimesterc	Maternal report of child sleep problems using sleep problems subscale of CBCL	4 y	Short prenatal sleep duration was not significantly associated with offspring sleep problems (all q values >0.12). Findings were only reported after adjusting for covariates.	Maternal age, preconception BMI, maternal tobacco/alcohol/substance use, prenatal depression and health problems, offspring sex, household income
Lyu et al. (2020)	6236 mother-child dyads in China Mage not reported continuously. 76.4% of the maternal sample was 25–34 y old	Self-reported sleep durationa	Retrospective report in early and mid-to-late pregnancy	Maternal report of child sleep habits using CSHQ	3–7 y	Short maternal sleep duration across pregnancy was associated with short sleep duration and more sleep disturbances in their offspring. Findings persisted after adjusting for covariates.	Maternal age and education, preconception weight, prenatal substance use and physical activity, maternal stressful life events, family structure, income, family history of sleep disorder, child age, sex, weight, allergies, attention-deficit/hyperactivity disorder, physical activity, and screen exposure
Morales-Muñoz et al. (2019)	777 mother-offspring dyads in Finland Mage = 30.6	Self-reported chronotype using MEQ	Once, mid-to-late pregnancy (M = 34.7 weeks’ GA, SD = 2.5)	Maternal report of offspring sleep quality using BISQ	3, 8, 18, and 24 mo	Mothers with an evening chronotype during pregnancy had offspring with short daytime sleep duration at 8 mo; short nighttime sleep duration at 3 and 8 mo; long sleep onset latency at 3, 18, and 24 mo; late bedtime at 3, 8, and 18 mo; and more sleep disturbances at 8 and 24 mo. Having a prenatal evening chronotype was not significantly associated with offspring total sleep duration (all b values < 0.06, all p values > .07) or night wakings (all b values < 0.06, all p values > .05) at any timepoint. Findings were only reported after adjusting for covariates.	Maternal age, gestational age at prenatal timepoint and at birth, parity, paternal age, child age and gender, season of birth, use of pacifier, and breastfeeding status
Nakahara et al. (2020) b	81,821 mother-infant dyads in Japan Mage not reported continuously. 35.4% of the maternal sample was 30–34 y old	Self-reported sleep duration and bedtimea	Once, mid-to-late pregnancyc	Maternal report of 1 mo infant night waking and whether infant slept longer during the day than the nighta	Short prenatal maternal sleep duration and late bedtime were associated with longer infant daytime sleep compared to nighttime sleep. Findings were only reported after covariates (model 1: only contained maternal age; model 2: all covariates) and persisted after adjusting for all covariates.	Maternal age, smoking habits and alcohol consumption, preconception BMI, parity, gestational age at birth, infertility treatment, type of delivery, infant sex, and postpartum depression symptoms
Nakahara et al. (2021) b	73,827 mother-infant dyads in Japan Mage not reported continuously. 36.0% of the maternal sample was 30–34 y old	Self-reported sleep duration and bedtimea	Once, mid-to-late pregnancy (25–35 weeks’ GA)	Maternal report 1 y of infant sleep habitsa	Short maternal sleep duration during pregnancy predicted a higher presence of night wakings, short sleep duration, and a late bedtime in the infant. Long prenatal maternal sleep duration (>10 h) predicted late infant bedtime. Additionally, prenatal maternal bedtime after midnight predicted more night wakings, short sleep duration, late bedtime, and bedtime resistance. Findings were only reported after covariates (model 1: only contained maternal age; model 2: all covariates) and persisted after adjusting for all covariates.	Maternal age, smoking habits and alcohol consumption, preconception BMI, parity, gestational age at birth, infertility treatment, and infant sex
Newland et al. (2016)	173 mother-infant dyads in the United States Mage = 29.0	Self-reported sleep duration and patternsa	Once, during pregnancy, timing not reportedc	Infant sleep efficiency 8 mo and variability measured by actigraphy; maternal report of infant night wakingsa	Short prenatal maternal sleep duration was correlated with high infant sleep efficiency and low variability in sleep efficiency. Prenatal maternal sleep duration was not significantly correlated with infant night wakings (all r values < 0.14).
Section II: Prenatal sleep quality
Armstrong etal. (1998)	97 mother-infant dyads in Australia Mage = 30.0	Self-reported sleep qualitya	Retrospective report of sleep early and mid-to-late pregnancy	Maternal report of infant sleep problemsa	1 y	Poor prenatal maternal sleep quality was associated with more infant sleep problems.
Cai et al. (2022)	797 mother-infant dyads in Singapore Afage = 30.7	Self-reported sleep quality using PSQI	Once, mid-to-late pregnancy (26–28 weeks’ GA)	Maternal report of infant sleep duration and wakefulness after sleep onset using BISQ	3, 6, 9, and 12 mo	Poor prenatal maternal sleep quality was associated with short infant nighttime and total sleep duration and long wakefulness after sleep onset. Poor sleep quality during pregnancy was not significantly associated with daytime sleep duration (p = .44) or number of night wakings (p = .35). Findings were only reported after adjusting for covariates.	Maternal age and education, parity, infant sex, and ethnicity
Cohen et al. (2022)	179 mother-child dyads in the United States Mage = 25.1	Self-reported sleep disturbances using PROMIS Sleep Disturbance Short Form	Once, mid-to-late pregnancy (M = 26.6 weeks’ GA, SD = 2.7)	Maternal report of child sleep habits using CSHQ	2y	Poor prenatal maternal sleep quality was correlated with poor offspring sleep quality. In mediation analyses, poor prenatal maternal sleep quality did not mediate the association between maternal discrimination and offspring sleep quality.	In mediation model, covariates were maternal age, socioeconomic status, prenatal substance use, postnatal depression symptoms, child age, and number of people in the home
Lin et al. (2022)	513 mother-infant dyads in China Mage = 29.0	Self-reported sleep quality using PSQI	Once, mid-to-late pregnancyc	Maternal report of infant sleep problems using BISQ	Within 3 mo of birth	Prenatal maternal sleep quality was not significantly associated with infant sleep outcomes (p = .75). However, prenatal maternal sleep problems that persisted in the postpartum were associated with a higher risk for infant sleep disorders.
Lugo-Candelas et al. (2023) b	794 mother-child dyads in the United States Mage = 32.3	Self-reported sleep quality and disturbances using MSHRa and/or PSQI	Three times, once in each trimesterc	Maternal report of child sleep problems using Sleep Problems subscale of CBCL	4 y	Greater maternal sleep disturbances in the second trimester were associated with more sleep problems in offspring. Prenatal sleep quality was not associated with child sleep problems (all q values > 0.07). Findings were only reported after adjusting for covariates.	Maternal age, preconception BMI, maternal tobacco/alcohol/substance use, prenatal depression and health problems, offspring sex, household income
Morales-Muñoz et al. (2018) b	1221 mother-infant dyads in Finland Mage = 30.6	Self-reported insomnia using BNSQ	Once, mid-to-late pregnancy (M = 34.7, SD = 2.5 weeks’ GA)	Maternal report of infant sleep difficulties using BISQ	3 mo	Poor prenatal maternal sleep quality was associated with short infant sleep duration and long sleep onset latency, but not significantly associated with nighttime self-soothing (p = .50), night wakings (p = .19), or irregular sleeping routine (p = .67). Findings were only reported after adjusting for covariates.	Maternal age, gestational age at prenatal timepoint, number of children in the family, infant age, season of birth, use of pacifier, and breastfeeding status
Nakahara et al. (2020) b	81,821 mother-infant dyads in Japan Mage not reported continuously. 35.4% of the maternal sample was 30–34 y old	Self-reported sleep quality based on sleep depth and feeling upon waking up in the past montha	Once, mid-to-late pregnancyc	Maternal report of infant night wakings and whether infants slept longer during the day than the nighta	1 mo	Poor prenatal maternal sleep quality was associated with more infant night wakings and long daytime sleep duration. Findings were only reported after covariates (model 1: only contained maternal age; model 2: all covariates) and persisted after adjusting for all covariates.	Maternal age, smoking habits, alcohol consumption, pre-pregnancy BMI, parity, gestational age at birth, infertility treatment, infant sex, type of delivery, small for gestational age, and postpartum depression symptoms
Nakahara et al. (2021) b	73,827 mother-infant dyads in Japan Mage not reported continuously. 36.0% of the maternal sample was 30–34 y old	Self-reported sleep quality based on sleep depth and feeling upon waking up in the past montha	Once, mid-to-late pregnancy (25–35 weeks’ GA)	Maternal report of infant sleep habitsa	1 y	Poor prenatal maternal sleep quality was associated with high prevelance of infant night wakings, short nighttime sleep duration, late bedtime, and higher frequency of crying at night. Findings were only reported after covariates (model 1: only contained maternal age; model 2: all covariates) and persisted after adjusting for all covariates.	Maternal age, smoking habits, alcohol consumption, pre-pregnancy BMI, gestational age at birth, infertility treatment, parity, and infant sex
Yang et al. (2018) b	169 mother-child dyads in China Mage not reported	Self-reported sleep problemsa	Retrospective report of entire pregnancyc	Maternal report of child gastrointestinal problems using a clinical questionnairea	5 y	Poor maternal sleep quality during pregnancy was associated with sleep problems in children with autism-spectrum disorder (ASD).
Section III: Prenatal sleep disorder or symptoms of disorders
Bin et al. (2017) b	626,188 mother-offspring dyads in Australia Mage not reported continuously. 68.0% of the maternal sample was 20–34 y old	Sleep apnea diagnosis via medical records	Once, diagnosis during pregnancy	Child hospitalizations in the first 6 y of life	0–9 y	Prenatal maternal sleep apnea diagnosis was associated with an increased risk of hospitalizations in offspring. Post-hoc analyses revealed that prenatal maternal sleep apnea predicted more sleep apnea-related hospitalizations in offspring.	Country of birth, maternal age, parental education and occupation, parity, delivery type, prenatal smoking, NICU admission, infant sex, birth weight, gestational age at birth, socioeconomic disadvantage, child age at assessment, non-English speaking background
Ciciolla et al. (2022)	59 mother-infant dyads in the United States Mage = 25.8	Self-reported insomnia symptoms using WHIIRS	Once, mid-to-late pregnancy (30 weeks’ GA)	Maternal report of infant sleep problems using ISQ	6 wk and 16 wk	Prenatal maternal insomnia symptoms were correlated with more infant sleep problems at 6 wk as compared to infants of mothers with no insomnia symptoms. However, prenatal maternal insomnia symptoms were not significantly correlated with infant sleep problems at 16 wk (r = −0.07). In path model, prenatal maternal insomnia symptoms predicted infant sleep problems at 6 wk through maternal postnatal insomnia.	In the path model, covariates were maternal race, education, infant sex
Tauman et al. (2015)	74 mother-infant dyads in Israel Mage = 32.9	Presence of sleep-disordered breathing using ambulatory sleep study	Once, mid-to-late pregnancy (33–36 weeks’ GA)	Maternal report of typical infant sleep patterns using BISQ	12 mo	Prenatal maternal sleep-disordered breathing predicted more infant snoring than infants of mothers without sleep-disordered breathing during pregnancy. Group differences (control versus sleep-disordered breathing group) do not significantly predict nighttime sleep duration (p = .07), night wakings (p = .69), sleep-onset latency (p = .14), wakefulness after sleep onset (p = .70), or perception of infant sleep (p = .85).	Maternal socioeconomic status, gestational age, birth weight, infant sex, and Apgar score at 5 min
Section IV: Prenatal daytime sleepiness
Lahti-Pulkkinen et al. (2018)	111 mother-child dyads in the United Kingdom Mage = 32.5	Self-reported daytime sleepiness using ESS	Twice, early (17 weeks’ GA) and mid-to-late pregnancy (28 weeks’ GA)e	Maternal report of child sleep problems using CSHQ	4 y	High prenatal maternal daytime sleepiness was associated with a higher presence of child sleep problems. Findings were only reported after adjusting for covariates.	Child age and sex
Morales-Muñoz et al. (2018) b	1221 mother-infant dyads in Finland Mige = 30.6	Self-reported sleepiness using ESS	Once, mid-to-late pregnancy (M = 34.7 weeks’ GA, SD = 2.5)	Maternal report of infant sleep difficulties using BISQ	3 mo	Prenatal maternal sleepiness was not significantly associated with infant sleep duration (p = .89), night wakings (p = .07), sleep onset latency (p = .83), or irregular sleeping routine (p = .99). Findings were only reported after adjusting for covariates.	Maternal age, gestational age at prenatal timepoint, number of children in the family, infant age, season of birth, use of pacifier, and breastfeeding status
(B) Studies With Offspring Physical Health Outcomes
Study	Sample Size, Country, and Maternal Age	Prenatal Sleep Measure	Frequency and Timing of Prenatal Assessment	Offspring outcome(s)	Age of Offspring Outcome(s)	Findings	Covariates
Section I: prenatal sleep timing
Chen et al. (2020)	6236 mother-offspring dyads in China Mage not reported continuously. 45.4% of the maternal sample was 25–29 y old	Self-reported sleep durationa	Retrospective report of entire pregnancyc	Presence of respiratory allergies	5 y	Short prenatal maternal nighttime sleep duration (<8 h/d) was associated with increased risk of respiratory allergies in offspring. When stratified by sex, findings were stronger for males than for females.	Model 1: Child age, child overweight/obesity status, child sleep duration, exercise frequency, and screen exposure, family income, family structure, maternal and paternal education, family history of allergies Model 2: Covariates in model 1 and gestational age at birth, type of delivery, breastfeeding status, maternal smoking and drinking, gestational hypertension, diabetes, and anemia, preconception overweight/obesity status, maternal age
Harskamp-van Ginkel et al. (2020)	3608 mother-child dyads in Greece and Netherlands Mage = 31.6	Self-reported daily sleep duration during the past montha	Once; early pregnancy in Dutch cohort (Mdn = 16 weeks’ GA) and mid-to-late pregnancy in Greek cohort (Mdn = 32 weeks’ GA)c	Weight, height, waist circumference, percentage of body fat, diastolic blood pressure, systolic blood pressure, BMI, and lipid profile	Greek cohort: 4 and 6 y Dutch cohort: 5, 10, and 11 y	Short prenatal maternal sleep duration was associated with higher child BMI and diastolic blood pressure, and larger waist circumference. Prenatal maternal sleep duration was not significantly associated with child percent body fat (p = .16) or lipid profile (p values > .69). When stratified by sex, these associations were significant for females, but not males. Findings were only reported after adjusting for covariates.	Maternal age, parity, prenatal smoking, preconception BMI, maternal education and country of origin, child sex and age
Meng et al. (2022) b	2211 mother-child dyads in China Mage = 28.5	Self-reported sleep duration and midpointa	Once, mid-to-late pregnancyc	BMI and subcutaneous fat (measured through triceps and subscapular skinfold thickness)	2y	Late prenatal maternal sleep midpoint predicted more subcutaneous fat in offspring. Short prenatal sleep duration was associated with higher offspring BMI. After adjusting for covariates, association between prenatal sleep midpoint and child subcutaneous fat remained significant, whereas the association between prenatal sleep duration and child BMI did not (p = .88).	Maternal age and education, preconception BMI, gestational weight gain, and infant sex
Mizutani et al. (2007)	1417 mother-child dyads in Japan Mage = 28.6	Self-reported sleep durationa	Once, early pregnancyd	Obesity/overweight status based on BMI	5 y	Prenatal maternal sleep duration was associated with childhood obesity and overweight status. Findings do not persist after adjusting for covariates	Maternal BMI and age
Zou et al. (2022) b	382 mother-offspring dyads in China Mage = 29.3	Sleep duration measured using a wearable device	Twice, early (16 weeks’ GA) and mid-to-late (34 weeks’ GA) pregnancy	BMI from body weight and length	1, 2, 4, 6, 9, 12, 18, and 24 mo	Inappropriate (short or long) maternal nighttime sleep duration in early pregnancy predicted higher risk of overweight or obesity in offspring. Increased maternal nighttime sleep duration from early to mid-to-late pregnancy predicted a reduced risk of the child developing overweightness or obesity in offspring. Findings were only reported after adjusting for covariates.	Preconception BMI, maternal gestational weight gain, maternal physical activity and energy intake, maternal anxiety and depression symptoms, obstetric complications, morning sickness in early pregnancy, feeding practices of first 6 mo, breastfeeding duration, gestational age at birth, and offspring sex
Section II: Prenatal sleep quality
Meng et al. (2022) b	2211 mother-child dyads in China Mage = 28.5	Self-reported sleep qualitya	Once, mid-to-late pregnancyc	BMI and subcutaneous fat (measured through triceps and subscapular skinfold	2y	Prenatal maternal sleep quality was not significantly associated with offspring adiposity indicators (p values > .23). Findings persisted after adjusting for covariates.	Maternal age and education, preconception BMI, gestational weight gain, and infant sex
Phelan et al. (2011)	285 mother-infant dyads in the United States Mage = 28.5	Self-reported sleep disturbances using GSDS	Once, mid-to-late pregnancy (30 weeks’ GA)	Child risk of obesity obtained from weight for age scores	6 mo	Prenatal maternal sleep quality was not significantly associated with infant weight (p value not reported). Findings persisted after adjusting for covariates.	Treatment group, infant sex, recruitment clinic, gestational age at birth, breastfeeding
Yang et al. (2018) b	169 mother-child dyads in China Mage not reported	Self-reported sleep problemsa	Retrospective report of entire pregnancyc	Maternal report of child gastrointestinal problems using a clinical questionnairea	5 y	Poor maternal sleep quality during pregnancy was associated with higher prevalence of gastrointestinal problems in children with autism-spectrum disorder (ASD).
Zou et al. (2022) b	3329 mother-offspring dyads in China Mage = 29.3	Self-reported sleep quality using PSQI	Twice, early (16 weeks’ GA) and mid-to-late (34 weeks’ GA) pregnancy	BMI from body weight and length	1, 2, 4, 6, 9, 12, 18, and 24 mo	Poor prenatal maternal sleep quality was associated with increased risk of overweight/obesity status. In the control group, improvement in maternal sleep quality from early pregnancy to late pregnancy was a protective factor against catch-up growth in offspring. Findings were only reported after adjusting for covariates.	Preconception BMI, maternal gestational weight gain, maternal physical activity and energy intake, maternal anxiety and depression symptoms, obstetric complications, morning sickness in early pregnancy, feeding practices of first 6 mo, breastfeeding duration, gestational age at birth, and offspring sex, and feeding practices
Section III: Prenatal sleep disorder or sleep disorder symptoms
Bin et al. (2017) b	626,188 mother-offspring dyads in Australia Mage not reported continuously. 68.0% of the maternal sample was 20–34 y old	Sleep apnea diagnosis via medical records	Once, diagnosis during pregnancy via medical records	Mortality rate and hospitalizations in the first 6 y of life	0–6 y	Diagnosis of prenatal sleep apnea was associated with higher prevalence of hospitalizations in infancy, including an elevated risk for pediatric sleep apnea hospitalizations. There was no significant association between prenatal sleep apnea and offspring mortality (p = .48). Findings persisted after adjusting for covariates.	Maternal country of birth, maternal age, residential zip code, prenatal substance use, parity, delivery type, gestational age at birth, NICU admission, infant sex, and birth weight
Brener et al. (2020)	58 mother-offspring dyads in Israel Afage = 33.5	Sleep-disordered breathing measured via apnea hypopnea index and Sp02	Once, mid-to-late pregnancy (33–36 weeks’ GA)	Length, weight, head circumference; subscapular and triceps skinfolds	0, 1, and 4 mo, and 1, 2, and 3y	Maternal sleep-disordered breathing during pregnancy predicted rapid catch-up head growth during the first year of life, as well as higher adiposity measured through triceps skinfolds in the first 3 y of life. Maternal sleep-disordered breathing during pregnancy did not significantly predict offspring weight (p = .61), length (p = .41), or subscapular skinfolds (p-value not reported). Findings were only reported after adjusting for covariates.	Gestational age at birth, maternal BMI, and paternal BMI
Felder et al. (2023)	2,963,029 mother-infant dyads in the United States Mage not reported continuously. 69.5% of the maternal sample was 18–34 y old	Prenatal sleep apnea or insomnia diagnosis via medical records	Once, diagnosis during pregnancy via medical records	Infant hospital admission, emergency départaient visit, and infant death	Within 1 y of age	Prenatal maternal sleep apnea and insomnia were associated with higher prevalence of infant emergency room visits, hospital stays, higher risk of mortality, and higher risk of hypoglycemia in the first year of life. After propensity score matching, prenatal maternal sleep apnea and insomnia were only associated with more emergency room visits in the first year of life.	Prenatal substance use, gestational hypertension, diabetes, and infection, preconception BMI, maternal education, history of preterm birth, enrollment in Women Infants and Children (WIC) program, payer for delivery, and birth weight

MSHP = Environmental Influences on Child Health Outcomes Study Maternal Sleep Health in Pregnancy Questionnaire; PSQI = Pittsburgh Sleep Quality Index; CBCL = Child Behavior Checklist Preschool Version; BMI = body mass index; CSHQ = Children’s Sleep Habits Questionnaire; MEQ = Morningness-Eveningness Questionnaire; GA = gestational age; BISQ = Brief Infant Sleep Questionnaire; PROMIS = Patient-Reported Outcomes Measurement Information System; BNSQ = Basic Nordic Sleep Questionnaire; NICU = neonatal intensive care unit; WHIIRS = Women’s Health Initiative Insomnia Rating Scale; ISQ = Infant Sleep Questionnaire; ESS = Epworth Sleepiness Scale; GSDS = General Sleep Disturbance Scale.

^aThe measure used was not a validated questionnaire.

^bStudy is presented in this table more than once.

^cNo specification on gestational weeks at the time of the prenatal sleep assessment was given.

^dParticipants were pooled across different gestational periods.

^eThe average of the scores obtained during multiple assessments was used in analyses.

TABLE 2.

Studies With Offspring Developmental and Brain Structure/Functlon Outcomes

(a) Studies With Offspring Developmental Outcomes
Study	Sample Size, Country, and Maternal Age	Prenatal Sleep Measure	Frequency and Timing of Prenatal Assessment	Offspring Outcome(s)	Age of Offspring Outcome(s)	Findings	Covariates
Section I: Prenatal sleep timing
Koutra et al. (2022) b	638 mother-child dyads in Greece Mage = 29.8	Self-reported sleep durationa	Once, mid-to-late pregnancy (28–32 weeks’ GA)	Maternal report of neuropsychological and behavioral development as measured by MSCA, ADHDT, and SDQ	4 y	Short prenatal maternal sleep duration was not significantly associated with neuropsychological and behavioral outcomes (all p values > .06). Findings were reported after adjusting for covariates (model 1: quality of assessment, examiner, and child sex; model 2: contained all covariates). After adjusting for all covariates, short prenatal sleep duration was associated with lower general cognition scores and reduced memory span in children.	Maternal education, maternal age at delivery, working during pregnancy, maternal prenatal snoring and daytime sleepiness, marital status, parity, prenatal maternal and paternal smoking, breastfeeding duration, preconception BMI, child sex, preschool attendance, passive smoking at the age of 4, quality of assessment, and examiner
Li et al. (2023)	544 mother-child dyads in Taiwan Mage = 33.6	Sleep duration patterns defined through self-reported sleep durationa	Three times, once in early pregnancy and twice in mid-to-late pregnancyc	Maternal report of offspring gross motor, language, and socio-emotional development using Taipei-II	1 and 6 mo, and 1, 2, and 3 y	Short decreasing and extremely long decreasing prenatal maternal sleep duration patterns were associated with overall child development. Extremely long decreasing patterns were associated with worse gross motor development Short decreasing, stably short, and extremely long decreasing sleep patterns were associated with poor language development. Prenatal sleep patterns were not significantly associated with child socioemotional development (all p values > .05). After adjusting for covariates, the association between short decreasing sleep patterns and language development became nonsignificant (p > .05), but all other findings persisted.	Maternal education, health and employment status, parental economic stress, depression, and anxiety during pregnancy, paternal age, and infant birth order, sex, and offspring reported health
Lugo-Candelas et al. (2023) b	794 mother-child dyads in die United States Mage = 32.3	Self-reported sleep duration using MSHPa and/or PSQI	Three times, once in each trimesterc	Maternal report of child ADHD symptoms and emotional reactivity using Attention Deficit/Hyperactivity-DSM5 and Emotional Reactivity subscales ofCBCL	4 y	Short prenatal sleep duration in the second trimester was associated with more child ADHD symptoms and greater emotional reactivity. Findings were only reported after adjusting for covariates.	Maternal age, preconception BMI, maternal tobacco/alcohol/substance use, prenatal depression and health problems, offspring sex, household income
Nakahara et al. (2020) b	81,821 mother-infant dyads in Japan Mage not reported continuously. 35.4% of the maternal sample was 30–34 y old	Self-reported sleep duration and bedtimea	Once, mid-to-late pregnancyc	Maternal report of infant temperament based on crying patternsa	1 mo	Prenatal maternal short sleep duration and late bedtime predicted frequent and intense crying in the infant. Findings were only reported after adjusting for some covariates.	Maternal age at delivery, smoking habits and alcohol consumption, preconception BMI, parity, gestational age at birth, infertility treatment, type of delivery, infant sex, small for gestational age, and postpartum depression symptoms
Nakahara et al. (2021) b	73,827 mother-infant dyads in Japan Mage not reported continuously. 36.0% of die maternal sample was 30–34 y old	Self-reported sleep duration and bedtimea	Once, mid-to-late pregnancy (25–35 weeks’ GA)	Maternal report of offspring development (e.g., problem-solving, motor, interpersonal skills) using ASQ	1 y	Prenatal maternal sleep duration and late bedtime were not significantly associated with child developmental outcomes (p values not reported). Findings were only reported after adjusting for covariates.	Maternal age, smoking habits and alcohol consumption, preconception BMI, parity, infertility treatment, gestational age at birth, and infant sex
Zhu et al. (2024)	7069 mother-child dyads in China Mage = 30.3	Self-reported sleep duration using PSQI (only sleep duration question)	Twice in mid-to-late pregnancy (24–28 and 32–36 weeks’ GA)	Maternal report of offspring motor, language, and personal-social development using DDST-II	6–36 mo	Prenatal maternal short sleep duration increased child risk of developmental delay. When compounded with gestational diabetes, risk for developmental delays increased further. Findings persisted after adjusting for covariates.	Maternal age, education level, prepregnancy BMI, blood pressure, anemia, depressive symptoms, family history of diabetes, physical activity, nutrition, paternal smoking and alcohol, short sleep duration in third trimester, delivery mode, infant sex, birth weight, prematurity, GA, breastfeeding, and hospitals
Section II: Prenatal sleep quality
Lugo-Candelas et al. (2023) b	794 mother-child dyads in the United States Mage = 32.3	Self-reported sleep quality and disturbances using MSHPa and/or PSQI	Three times, once in each trimesterc	Maternal report of child ADHD symptoms and emotional reactivity using Attention Deficit/Hyperactivity-DSM5 and Emotional Reactivity subscales of CBCL	4 y	Poor prenatal maternal sleep quality in the second trimester and greater maternal sleep disturbances in the first trimester predicted more child ADHD symptoms. More maternal sleep disturbances in the third trimester were associated with greater child emotional reactivity, whereas prenatal sleep quality was not (all q values > 0.48). Findings were only reported after adjusting for covariates.	Maternal age, preconception BMI, maternal tobacco/alcohol/substance use, prenatal depression and health problems, offspring sex, household income
Nakahara et al. (2020) b	81,821 mother-infant dyads in Japan Mage not reported continuously. 35.4% of the maternal sample was 30–34 y old	Self-reported sleep quality based on sleep depth and feeling upon waking upa	Once, mid-to-late pregnancyc	Maternal report of infant temperament based on crying patternsa	1 mo	Poor prenatal maternal sleep quality was associated with infant frequent and intense crying. Findings were reported after adjusting for covariates.	Maternal age at delivery, smoking habits and alcohol consumption, preconception BMI, parity, gestational age at birth, infertility treatment, type of delivery, infant sex, small for gestational age, and postpartum depression symptoms
Nakahara et al. (2021) b	73,827 mother-infant dyads in Japan Mage not reported continuously. 36.0% of the maternal sample was 30–34 y old	Self-reported sleep quality based on sleep depth and feeling upon waking up in the past montha	Once, mid-to-late pregnancy (25–35 weeks’ GA)	Maternal report of offspring developmental outcomes (e.g., communication, problem-solving, motor, and interpersonal skills) using ASQ	1 y	Poor prenatal maternal sleep quality was associated with an increased risk of compromised development on all domains of the ASQ. Finding only reported after adjusting for covariates.	Maternal age, smoking habits and alcohol consumption, preconception BMI, parity, gestational age at birth, infertility treatment, and infant sex
Section III: Prenatal sleep disorders and/or symptoms
Adler et al. (2021)	1346 mother-child dyads in Norway Mage = 31.7	Self-reported insomnia symptoms using BIS	Once, mid-to-late pregnancy (32 weeks’ GA)	Maternal report of social and emotional development using the ASQ	2y	Prenatal maternal insomnia predicted higher prevalence of social-emotional problems in their offspring. Findings persisted after adjusting for covariates.	Maternal age, education, work status, depression symptoms, breastfeeding status, child sex, and preterm birth
Bin et al. (2017)	626,188 mother-offspring dyads in Australia Mage not reported continuously. 68.0% of die maternal sample was 20–34 y old	Sleep apnea diagnosis via medical records	Once, diagnosis during pregnancy	Developmental vulnerability, special needs diagnosis, and standardized test performance in reading and numeracy using AvEDI and NAPLAN	0–9 y	Prenatal maternal sleep apnea was not significantly associated with developmental vulnerability (p = .48), special needs status (p = .30), or reading (p = .09) and numeracy scores (p = .72). Sex-specific analyses revealed that prenatal apnea predicted developmental vulnerability in males but not females. After adjusting for covariates, prenatal sleep apnea predicted lower reading scores.	Maternal age, maternal country of birth, socioeconomic disadvantage, parental occupation, parity, year of child assessment, age at test, and non-English-speaking background
Ciciolla et al. (2022)	59 mother-child dyads in the United States Mage = 25.8	Self-reported insomnia symptoms using WHIIRS	Once, mid-to-late pregnancy (30 weeks’ GA)	Maternal report of infant negative reactivity, orienting, and surgency using the IBQ-R	16 wk	Prenatal maternal insomnia was not significantly correlated with infant negative emotionality (r = 0.31), orienting (r = 0.03), and surgency (r = 0.28). In die path model, prenatal maternal insomnia predicted infant negative reactivity at 16 wk through postnatal maternal insomnia symptoms and infant sleep difficulties at 6 wk.	In the path model, covariates were maternal race, education, infant sex, infant sleep problems at 16 wk
Koutra et al. (2022) b	638 mother-child dyads in Greece Mage = 29.8	Self-reported snoringa	Once, mid-to-late pregnancy (28–32 weeks’ GA)	Maternal report of neuropsychological and behavioral development as measured by MSCA, the ADHDT, and SDQ	4 y	Prenatal maternal snoring was not associated with neuropsychological and behavioral development outcomes (all p values > .06). Findings were reported after adjusting for covariates (model 1: quality of assessment, examiner, and child sex; model 2: contained all covariates). After adjusting for all covariates, prenatal maternal snoring predicted more hyperactivity.	Maternal education, maternal age at delivery, working during pregnancy, maternal prenatal daytime sleepiness and sleep duration, marital status, parity, prenatal maternal and paternal smoking, breastfeeding duration, and preconception BMI, child sex, preschool attendance, passive smoking at the age of 4, quality of assessment and examiner
Morrakotkhiew et al. (2021)	159 mother-offspring dyads in Thailand Mage = 33.8	Presence and severity of obstructive sleep apnea determined by a two-night polysomnography	Once, early pregnancy (0–16 weeks’ GA)	Maternal report of early childhood development/risk of developmental delay using ASQ	6–36 mo	Prenatal maternal obstructive sleep apnea was associated with increased risk for developmental delay in offspring. Findings persisted after adjusting for covariates.	Birth asphyxia, breastfeeding status, screen time, maternal education and income, hypertension, gestational diabetes mellitus, and obesity
Tauman et al. (2015)	74 mother-infant dyads in Israel Mage = 32.9	Presence of sleep-disordered breathing using several respiratory indicesa	Once, mid-to-late pregnancy (33–36 weeks’ GA)	Maternal report of infant development level using DIQ and researcher-observed spontaneous general movementsa	12 mo	Prenatal maternal sleep-disordered breathing was not significantly associated with social (p = .07), self-help (p = .93), gross motor (p = .95), fine motor (p = .95), language skills (p = .74) or spontaneous general movements (p = .19) in the infant. Findings persisted after adjusting for covariates.	Maternal socioeconomic status, gestational age, birth weight, infant sex, and Apgar score at 5 min
Vizzini et al. (2018)	3634 mother-child dyads in Italy Mage not reported continuously 43.3% of the maternal sample was 30–34 y old	Presence of any diagnosed sleep disorder	Once, mid-to-late pregnancy (M= 26 weeks’ GA)	Maternal report of child ADHD symptoms based on the criteria for diagnosis in the DSM-V	4 and 7 y	Maternal prenatal diagnosis of sleep disorders was associated with an increase in the hyperactivity and impulsivity scores of the ADHD scales in the offspring.	Maternal age and education, prenatal maternal substance use, child gender, parity, and ADHD inattentive score
Section IV: Prenatal daytime sleepiness
Koutra et al. (2022) b	638 mother-child dyads in Greece Mage = 29.8	Self-reported daytime sleepiness using ESS	Once, mid-to-late pregnancy (28–32 weeks GA)	Maternal report of neuropsychological and behavioral development as measured by MSCA, the ADHDT, and SDQ	4 y	Prenatal maternal daytime sleepiness did not significantly predict neurop sychological and behavioral development outcomes (all ps > .08). Findings were reported after adjusting for covariates (model 1: quality of assessment, examiner, and child sex; model 2: contained all covariates). After adjusting for all covariates, prenatal maternal daytime sleepiness was associated with offspring poor memory.	Maternal education, maternal age at delivery, working during pregnancy, maternal prenatal snoring and sleep duration, marital status, parity, prenatal maternal and paternal smoking, breastfeeding duration, and preconception BMI, child sex, preschool attendance, passive smoking at the age of 4, quality of assessment and examiner
Lahti-Pulkkinen etal. (2018)	111 mother-child dyads in the United Kingdom Mage = 32.5	Self-reported daytime sleepiness using ESS	Twice, early (17 weeks GA) and mid-to-late pregnancy (28 weeks GA)e	Maternal report of child development using SDQ, CBCL, Connor’s Hyperactivity Scale, ASQ, HTKS, and Marshmallow Test	4 y	Prenatal maternal daytime sleepiness predicted more internalizing, externalizing, and total psychiatric difficulties, hyperactivity, poor neurodevelopment and executive functioning scores. Prenatal maternal daytime sleepiness did not significantly predict delay of gratification via the Marshmallow Test (p = .53). Results were only reported after adjusting for covariates.	Child age and sex
(b) Studies With Offspring Brain Structure and Function
Study	Sample Size, Country, and Maternal Age	Prenatal Sleep Measure	Frequency and Timing of Prenatal Assessment	Offspring Outcome(s)	Age of Offspring Outcome(s)	Findings	Covariates
Section I: Prenatal sleep timing
Hoyniak et al. (2023)	148 mother-infant dyads in the United States Mage = 29.8	Sleep duration, variability in duration, midpoint, and circadian misalignment measured using a wearable device	Three times, once in early pregnancy and twice in mid-to-late pregnancyc,e	Offspring brain structure (volumetric measures of white matter, cortical and subcortical gray matter, and cortical surface area) using structural MRI	First month of life	Greater night-to-night variability in sleep duration during pregnancy predicted reduced volumes in cortical and subcortical gray matter, white matter, and reduced cortical surface area. Later prenatal sleep midpoint and greater circadian misalignment predicted reduced infant subcortical gray matter volume but were not associated with other outcomes (all p values > .12). Prenatal sleep duration (all p values > .38) did not predict infant outcomes. Findings were only reported after adjusting for covariates.	Maternal age, preconception obesity, parity, offspring p ostconceptional age and sex
Section II: Prenatal sleep quality
Hoyniak et al. (2023)	148 mother-infant dyads in the United States Mage = 29.8	Sleep efficiency (i.e., sleep activity density) measured using a wearable device	Three times, once in early pregnancy and twice in mid-to-late pregnancyc,e	Offspring brain structure (volumetric measures of white matter, cortical and subcortical gray matter, and cortical surface area) using structural MRI	First month of life	Prenatal sleep efficiency (all p values > .10) did not significant predict infant outcomes. Findings were only reported after adjusting for covariates.	Maternal age, preconception obesity, parity, offspring p ostconceptional age and sex
Lavonius et al. (2020) b	142 mother-infant dyads in Finland Mage = 30.8	Self-reported sleep quality using BNSQ	Once, mid-to-late pregnancy (24 weeks’ GA)	Event-related potentials using EEG	4 wk	Poor prenatal maternal sleep quality predicted smaller infant ERP responses to happy stimuli. The association persisted after adjusting for covariates.	Maternal depression symptoms, maternal age and BMI, parity, neonatal gestational age at assessment, and sex
Nevarez-Brewster et al. (2022)	94 mother-infant dyads in the United States Mage = 30.5	Self-reported sleep quality using PSQI	Three times, once in early pregnancy (M = 16 weeks’ GA) and twice in mid-to-late pregnancy (M = 28 weeks’ GA; M= 35 weeks’ GA)	Newborn bilateral hippocampal and amygdala volume assessed through structural MRI	5 weeks	Poor maternal sleep quality was associated with large newborn hippocampal volume bilaterally. Poor maternal sleep quality was not associated with newborn amygdala volume. Findings were only reported after adjusting for covariates.	Prenatal income, birth weight percentile, intracranial volume, postconceptional age at scan, infant sex
Section III: Prenatal daytime sleepiness
Lavonius et al. (2020) b	142 mother-infant dyads in Finland Mage = 30.8	Self-reported sleepiness using BNSQ	Once, mid-to-late pregnancy (24 weeks’ GA)	Event-related potentials using EEG	4 wk	Prenatal maternal daytime sleepiness predicted smaller infant ERP responses to happy stimuli and higher ERP responses to sad stimuli. After adjusting for covariates, the association between prenatal maternal daytime sleepiness and happy stimuli persisted, whereas the effect on ERP responses to sad stimuli did not (p = .17).	Maternal depression symptoms, maternal age and BMI, parity, neonatal gestational age at assessment, and sex

GA = gestational age; MSCA = McCarthy Scale of Children’s Abilities; ADHDT = Attention-DeficifHyperactivity Disorder Test; SDQ = Strengths and Difficulties Questionnaire; BMI = body mass index; Taipei-II = Taipei City Developmental Screening Checklist for Preschoolers, 2nd Version; MSHP = Environmental Influences on Child Health Outcomes Study Maternal Sleep Health in Pregnancy Questionnaire; PSQI = Pittsburgh Sleep Quality Index; ADHD = attention-deficit/hyperactivity disorder; DSM5 = Diagnostic and Statistical Manual of Mental Disorders, 5th Edition; CBCL = Child Behavioral Checklist; ASQ=Ages and Stages Questionnaire; DDST-II = Denver Developmental Screening Test-II; GDM = gestational diabetes mellitus; BIS = Bergen Insomnia Scale; AvEDI = Australian version of the Early Development Instrument; NAPLAN = National Assessment Program-Literacy and Numeracy; WHIIRS = Women’s Health Initiative Insomnia Rating Scale; IBQ-R = Infant Behavior Questionnaire—Revised; DIQ = Developmental Inventory Questionnaire; ADHD = attention-deficit hyperactivity disorder; DSM-V = Diagnostic and Statistical Manual Version V; ESS = Epworth Sleepiness Scale; CBCL = Child Behavioral Checklist; HTKS = Heads, Toes, Knees, and Shoulders test; BNSQ = Basic Nordic Sleep Questionnaire; EEG = electroencephalogram; ERP = event-related potential; MRI = magnetic resonance imaging.

^aThe measure used was not a validated questionnaire.

^bStudy is presented in this table more than once.

^cNo specification on gestational weeks at the time of the prenatal sleep assessment was given.

^dParticipants were pooled across different gestational periods.

^eThe average of the scores obtained during multiple assessments was used in analyses.

Offspring Sleep

Half of the studies in this review (n = 16; 47.1%) examined associations between prenatal maternal sleep and offspring sleep health. Offspring sleep was most frequently examined during infancy (n = 10), with a few studies in early childhood (n = 4), infancy and early childhood (n = 1), and early to middle childhood (n = 1). Offspring sleep was assessed via parent-report, with the exception of one study that used actigraphy (55).

Overall, prenatal maternal sleep timing was associated with multiple indicators of maternal-reported offspring sleep (Table 1A, Section I). Late prenatal maternal bedtime was associated with short nighttime sleep duration, more daytime sleep compared to nighttime sleep, late bedtime, and more night wakings in infants, as well as high frequency of crying at night (56,57). Short prenatal maternal sleep duration also predicted offspring sleep problems, including short sleep duration (nighttime and total; (52,56)), more sleep disturbances (52,53), more daytime sleep compared to nighttime sleep (56,58), late bedtime, and more night wakings (57). Only one study implemented objective measures of infant sleep and found that shorter prenatal maternal sleep duration was linked to high infant sleep efficiency and low variability in infant sleep efficiency (55). Infants of mothers who had an evening chronotype during pregnancy had short daytime and nighttime sleep during infancy; long sleep onset latency at 18 and 24 months; late bedtime at 3, 8, and 18 months; and more sleep difficulties at 8 and 24 months compared to infants of mothers with morning chronotypes (59).

Prenatal maternal sleep quality was related to similar indices of maternal-reported offspring sleep as prenatal sleep timing (Table 1A, Section II). Poor prenatal sleep quality was associated with more offspring sleep problems in 1- to 5-year-olds (41,50,53,60), as well as short sleep duration within the first year of life (nighttime and total; (25,57,61)), long sleep onset latency at 3 months of age (61), more night wakings at 1 year of age (57), longer duration of wake after initial sleep onset across the first year of life (25), more daytime sleep compared to nighttime sleep, late bedtime, and high frequency of crying at night in 1-month and 1-year-olds (56,57). Poor prenatal sleep quality also predicted an increased risk of sleep problems within the first 6 months of life if poor maternal sleep extended from third trimester to the postpartum period (62). The studies from this subsection (Table 1A, Section II) were predominantly conducted prior to the offspring’s second birthday.

Sleep disorders and/or symptoms of sleep disorders and daytime sleepiness were also linked to offspring sleep (Table 1A, Sections III and IV). Prenatal insomnia symptoms were correlated with more infant sleep difficulties at 6 weeks but not at 16 weeks (40). The presence of sleep-disordered breathing during pregnancy was associated with higher risk of infant snoring but not with any other indicators of infant sleep at 12 months of age (63). Prenatal sleep apnea was associated with increased risk of hospital admissions for pediatric sleep apnea in the first year of life (64). Furthermore, more daytime sleepiness during pregnancy predicted greater sleep problems at 4 years old (65) but not during infancy (61), suggesting the need for studies extending to early childhood and adolescence in this area of research.

Of note, several of the studies that reported significant associations for some offspring sleep indicators also reported null associations for other sleep indicators (55,61,63). For these studies, the pattern of associations was not consistent. Differences in measurement and timing of prenatal and offspring sleep likely contribute to these discrepancies.

Offspring Physical Health

Physical health indicators (n = 10 studies; 29.4%) included anthropometrics (e.g., height, weight, body mass index [BMI], measures of adiposity; n = 6), presence of health conditions (allergies, respiratory distress syndrome, hypoglycemia; n = 3), mortality (n = 2), hospital visits (n = 2), blood pressure (n = 1), and lipid profiles (n = 1). Anthropometric measurements, blood pressure, and lipid profiles were obtained via trained medical or research staff; health conditions were collected via parent-report and allergen tests; and data on mortality and hospitalizations were from hospital records.

Prenatal sleep timing and quality were associated with offspring anthropometric outcomes and presence of health conditions (Table 1B, Sections I and II). Short prenatal maternal sleep duration was associated with higher risk of obesity/overweight status during infancy and early childhood (66,67); higher BMI, waist circumference, and diastolic blood pressure in early to middle childhood, and higher risk of respiratory allergies at 5 years of age (51,67,68). Prenatal sleep duration was not associated with child percent body fat or lipid profiles across two cohorts with repeated assessments of offspring physical outcomes from 4 to 11 years of age (67). Late maternal sleep midpoint predicted higher offspring subcutaneous fat at 2 years of age (69). Poor prenatal sleep quality predicted higher BMI across the first 2 years of life (51) but did not significantly predict other adiposity or growth indicators at 6 months or 2 years of age (69,70). The discrepancy in findings may be due to variation in prenatal maternal sleep assessment and timing of measurement offspring physical health. To illustrate, although offspring physical health was predominantly assessed objectively (e.g., medical charts, anthropometric assessment), the number of offspring physical health assessments ranged from one to eight across studies. Poor prenatal sleep quality was also associated with more gastrointestinal symptoms in a sample of 5-year-old children with autism spectrum disorder (60).

In terms of prenatal maternal sleep disorders (Table 1B, Section III), offspring of mothers with sleep-disordered breathing showed more rapid catch-up growth in head circumference and greater adiposity (measured via triceps skinfolds) than offspring of mothers without sleep-disordered breathing (71). In a large study of almost 3 million participants, offspring of mothers with diagnoses of prenatal insomnia or sleep apnea were at higher risk of respiratory distress, hypoglycemia, and mortality within the first 3 months of life, as well as more emergency room visits and greater risk of hospital admission within the first year of life, as compared to offspring of mothers without these disorders (72). However, after propensity score matching, prenatal sleep disorders were only associated with greater emergency room visits in the first year of life (72). Another large-scale study found that offspring of mothers with prenatal sleep apnea were more likely to be admitted to the hospital between birth to 6 years than offspring of mothers without sleep apnea (64).

Offspring Developmental Outcomes

Thirteen studies (38.2%) examined child developmental outcomes, including measures of global development and developmental delays (n = 9), attention-deficit/hyperactivity disorder (ADHD) symptoms (n = 4), behavioral problems (n = 2), negative affect (n = 2), executive function (EF; n = 1), reading and numeracy skills (n = 1), and general movements in the first 16 weeks of life (n = 1). These outcomes were mostly assessed via parent questionnaires, such as the Ages and Stages Questionnaire (ASQ; (73); n = 4) and Strengths & Difficulties Questionnaire (SDQ; (74); n = 2). Three studies used objective methods including EF tasks, school records, and observations of spontaneous infant movements.

Prenatal sleep timing and quality were mostly associated with measures of socioemotional functioning and global development, with a few exceptions (Table 2A, Sections I and II). Short sleep duration during pregnancy predicted higher offspring ADHD symptoms and greater risk of developmental delay in the first year of life (53,75). However, prenatal sleep duration did not significantly predict poor development in one study of 4-year-olds (57). In another study, short sleep duration was only associated with reduced cognitive and memory scores in 4-year-olds after adjusting for covariates (76). Interestingly, Li et al. (54) examined trajectories of prenatal sleep duration and found that short sleep duration across pregnancy was associated with a higher risk of overall developmental delay across the first 3 years of life. They also found that short sleep in the beginning of pregnancy with decreasing duration as pregnancy progressed was associated with a higher risk of language delay. Additionally, sleep patterns that were long (i.e., 9–10 hours per night) and shortened over gestation were associated with higher risk of overall developmental, gross motor, and language delays. Short duration, late bedtime, and poor sleep quality during pregnancy were associated with frequent and intense offspring crying in infancy, and more difficulty holding infants due to fussiness (56). Poor prenatal sleep quality also predicted poorer development as measured by low ASQ scores at 1 year of age (57) and higher ADHD symptoms at 4 years of age (53).

Findings from studies of sleep disorders and/or symptoms varied based on the type of disorder (Table 2A, Section III). First, the presence of any sleep disorder during pregnancy was associated with increased ADHD symptoms in offspring (77). Prenatal maternal snoring also predicted more ADHD hyperactivity symptoms, only after adjusting for covariates (76). The presence of prenatal sleep apnea was associated with poorer development (78) and lower reading scores, but only after adjusting for covariates (64). Prenatal insomnia symptoms were associated with offspring socioemotional problems (79), and also indirectly predicted greater infant negative reactivity (40). However, the presence of prenatal sleep-disordered breathing did not significantly predict social, motor, or language outcomes (63).

Finally, more prenatal maternal daytime sleepiness (Table 2A, Section IV) predicted more behavioral problems and poorer performance on one EF task of children’s working memory and inhibitory control (65). Daytime sleepiness also predicted poorer offspring memory, but only after adjusting for covariates (65). Both studies were conducted with children that were 4 years old and using self-report of offspring developmental outcomes.

Brain Structure and Function

Three studies (8.8%) examined how prenatal sleep was associated with offspring brain structure and function (Table 2B, Sections I–III). Poor sleep quality across pregnancy, and particularly during early pregnancy, predicted larger newborn hippocampal volume, but did not predict amygdala volume (49). Additionally, greater variability in prenatal maternal sleep duration across days presaged smaller total cortical gray and white matter volumes and reduced cortical surface areas in infants, whereas higher levels of circadian misalignment and later sleep timing during pregnancy were linked to smaller subcortical gray matter volumes (80). In terms of brain function, poor prenatal maternal sleep quality and more daytime sleepiness were associated with smaller event-related potential (ERP) responses to happy stimuli and higher ERP responses to sad stimuli (81), though the associations with responses to sad stimuli did not hold after adjusting for covariates. All the studies in this subsection were conducted within 5 weeks of birth, highlighting the need for studies assessing the links between prenatal maternal sleep and offspring brain structure and function beyond this developmental timepoint.

Risk of Bias

Overall risk of bias scores were calculated by summing each quality assessment category as follows: 1 = green, 2 = yellow, and 3 = red. The sum was then divided by four (the number of quality categories) for a possible score from 1 to 4. After rounding, studies were identified as “high quality” if they scored a 1, “moderate quality” if they scored a 2, or “low quality” if they scored a 3 or 4. Based on our criteria, 8.8% of studies included in this review were of high quality overall, 70.6% studies were of moderate quality, and 20.6% studies were of low quality (Figure 2). There was substantial heterogeneity in the measurement of prenatal sleep, with studies equally distributed across high (29.4%), moderate (35.3%), and low (35.3%) levels of measurement quality. The quality of offspring outcomes measures was, on average, higher than the quality of prenatal sleep measurement, with more than half of the studies (52.9%) using high-quality measures. Approaches to address missing data varied, with only 29.4% of studies using statistical methods like multiple imputation or maximum likelihood and more than half (55.9%) of studies using listwise deletion or not reporting missing data. Finally, almost half of the studies (47.1%) included covariates from two or three of the categories (sociodemographic risk, physical and mental health), whereas 23.5% of studies included covariates from none or included covariates only pertaining to offspring growth or maturation (e.g., gestational age at birth, child sex).

FIGURE 2.

Study Quality and Risk of Bias Stoplight. Color image is available online only at the journal website.

DISCUSSION

Sleep problems during pregnancy are pervasive, and this systematic review provides evidence for intergenerational consequences for offspring health and development. We identified associations among poor prenatal maternal sleep and some aspects of offspring sleep health (e.g., night wakings, sleep duration), physical health (e.g., BMI, blood pressure), developmental outcomes (e.g., behavioral problems), and brain structure and function (e.g., hippocampal volume). This review found critical gaps within the prenatal maternal sleep and offspring outcomes literature including reliance on self-report of sleep at a single timepoint, and lack of consideration of development after the age of 11. The present review provides recommendations for future work.

Associations Between Prenatal Sleep and Offspring Outcomes

The most studied outcome is offspring sleep. Existing evidence suggests that maternal sleep timing (shorter sleep duration, later maternal bedtime) and poor sleep quality during pregnancy predict offspring sleep disturbances. Various prenatal maternal sleep indicators and offspring sleep outcomes have been assessed across separate studies, and future research is needed investigating multiple prenatal sleep indicators and offspring outcomes simultaneously. Additionally, only one study assessed offspring sleep objectively via wearable devices (55). Thus, the field can benefit from more studies that use both parent-report and objective measures of offspring sleep to clarify the potential influence of prenatal sleep.

Prenatal maternal sleep (timing and quality) also predicts several offspring physical health outcomes, including early markers of cardiovascular health (e.g., BMI, blood pressure, and waist circumference). More severe sleep disruptions, including prenatal maternal sleep disorders or symptoms of disorders, are linked to more hospital-related outcomes in the first years of life (64,72). Future research examining other early indicators of cardiovascular health such as offspring biomarkers (e.g., glucose regulation, inflammation), satiety, and metabolism is needed.

A few studies evaluate the associations among prenatal maternal sleep and offspring developmental outcomes. Specifically, prenatal sleep disorders and worse sleep quality presage the presence of neurodevelopmental impairments in the offspring, including ADHD symptoms (53,77). Further, short sleep duration late in pregnancy and decreasing sleep duration across pregnancy may be associated with higher offspring irritability and cognitive deficiencies (57,59,60). The present findings are supported by one or two studies for each offspring outcome, and more studies are needed to replicate these associations.

Experimental rodent work indicates that maternal sleep is causally linked to changes in offspring neurocircuitry, specifically increases and decreases in neurogenesis in limbic regions, such as the hippocampus (82,83). There are very few imaging studies in humans, and our team recently demonstrated that poor prenatal maternal sleep predicted larger hippocampal volume in newborns (49). Our findings are consistent with the argument that while adapting to the demands of the immediate intrauterine environment, times of developmental plasticity are altered and maturation is accelerated, which may increase vulnerability to poor long-term offspring outcomes (84,85). Furthermore, the only other magnetic resonance imaging study in humans links prenatal maternal circadian misalignment and a later bedtime with lower subcortical gray matter in neonates (80). Future research should therefore evaluate the longer lasting neurodevelopmental consequences of prenatal sleep disruptions.

Strengths and Limitations

The studies included possess key strengths and limitations. Most studies (n = 29; 85.3%) were conducted prospectively, which reduces reporting bias in the postpartum period. A limitation of existing research is that most studies (n = 26; 76.5%) assessed prenatal sleep only once during pregnancy, and thus, it is not possible to look at differences related to timing of exposure during pregnancy. Another strength is the competitive sample sizes of these studies. More than half of the studies (n = 22; 64.7%) included at least 500 participants and included samples from across the globe. However, generalizability of the existing literature is limited as studies from regions in and around Africa, and Central and South America are lacking, as are studies with diverse samples. Offspring outcomes were assessed using multiple assessment modalities, including parent-report, medical charts, medical diagnoses, biological samples, and magnetic resonance imaging bolstering the argument that prenatal maternal sleep affects child outcomes. However, only a handful of studies collected repeated assessments of offspring outcomes, limiting the understanding of links between prenatal sleep and offspring maturation. Further, offspring outcomes did not extend past age 11 and predominantly focused on developmental outcomes prior to the age of five. Incorporating multiple assessments of prenatal maternal sleep and offspring development, and extending the age at offspring outcome assessment, would shed light on how sleep changes across gestation and how offspring growth is affected by trajectories of prenatal maternal sleep.

Future Directions

We overview several avenues for future research (see Supplemental Digital Content Table S4 for a summary, http://links.lww.com/PSYMED/B50).

Multidimensional Prenatal Sleep Health

Sleep health comprises numerous sleep dimensions, including but not limited to, nighttime sleep duration, sleep efficiency, sleep quality, bedtime and wake time, and daytime dysfunction. Various aspects of sleep have different implications for offspring development. In the current review, we found that prenatal sleep duration and quality were differentially associated with offspring outcomes. Future studies that implement a multidimensional approach to examining prenatal sleep (i.e., utilizing previously validated measures of sleep health, assessing sleep via wearable devices and polysomnography, and implementing several measurement modalities to understand how each inform offspring outcomes) would further our understanding of prenatal maternal sleep and how to improve it (86).

Timing and Frequency of Prenatal Sleep Assessment

It is unknown if there are sensitive periods during gestation when sleep problems are especially salient for offspring health. Evidence demonstrates that sleep changes robustly across gestation (17,47,87) and emerging cross-species evidence suggests that the consequences of poor sleep vary based on gestational timing (49,52,53,83). Further, it is plausible that patterns of sleep over gestation have implications for fetal maturation. Three studies found that a worsening of prenatal sleep from early to late pregnancy predicts poor physical health and changes in brain structure (49,51,54). Understanding how patterns of poor prenatal maternal sleep associate with infant outcomes would inform data-driven interventions to target specific prenatal timepoints in which sleep improvements may prove more beneficial for mother and child.

Exploring and Identifying Potential Mechanisms

Multiple pathways, including maternal cortisol and melatonin, maternal inflammation, offspring epigenetic modifications (e.g., deoxyribonucleic acid [DNA] methylation), and offspring neurocircuitry, may play a role in the intergenerational transmission of health and disease. We highlight a subset of plausible pathways here.

Cortisol and melatonin, key indices of circadian regulation, are affected by maternal sleep (29,88) and associate with infant outcomes (89,90), and thus are two potential signals by which prenatal maternal sleep may impact the fetus (22,31,91). Poor prenatal maternal sleep is linked to a flattened diurnal pattern of prenatal maternal cortisol (88) and heightened serum cortisol (92), which have been associated with poorer offspring outcomes (93,94). Similarly, poor sleep is associated with changes to maternal melatonin, and melatonin is related to offspring outcomes (95,96). Given the essential roles of cortisol and melatonin as circadian regulators and emerging evidence that prenatal maternal sleep is associated with maternal cortisol and melatonin (22,88), these neuroendocrine hormones likely play key roles in the association between prenatal maternal sleep and offspring outcomes.

Maternal inflammatory markers may also play a role in the link between prenatal sleep and offspring health. In pregnant individuals, greater subjective sleep disruptions are associated with higher levels of inflammatory cytokine IL-6 (28), IL-8 (19), and TNF-α (97), and maternal inflammation is additionally related to offspring health and development (98,99). Given preliminary evidence linking poor prenatal maternal sleep to maternal inflammation, future research investigating maternal inflammation as a pathway is warranted.

Prenatal maternal sleep may also affect fetal brain maturation, subsequently affecting offspring outcomes. Experimental rat studies have identified a causal link between maternal gestational sleep deprivation and alterations to offspring neurocircuitry (82,83). Evidence is now emerging indicating that prenatal maternal sleep problems also predict neonatal brain structure and function in humans (49,80), and these changes in offspring neurocircuitry are linked to later offspring cardiometabolic and socioemotional health in humans (100,101). Based on these findings, future research examining offspring neurocircuitry as a pathway linking prenatal maternal sleep and offspring outcomes is needed. In line with this recommendation for future research, one recent investigation from our group revealed that prenatal maternal sleep problems were associated with neonatal white matter maturation of the uncinate fasciculus, a major frontolimbic tract (102). Further, changes in uncinate fasciculus development were indirectly associated with increases in infant negative emotionality (measured multimodally).

Adolescent Outcomes

Although the present review sought to include studies examining the associations among poor prenatal maternal sleep and offspring outcomes through age 18, existing studies of prenatal sleep have not evaluated offspring outcomes beyond 11 years of age. Adolescence is a window of life when cognitive and emotional problems often emerge and understanding links with prenatal sleep could inform our understanding of intergenerational transmission of mental health.

Assessment of Sex Differences

Research exploring whether sex moderates the associations between poor prenatal maternal sleep and offspring outcomes is scant. Only six studies in the current review assessed sex differences in the associations between poor prenatal sleep and offspring outcomes. This small literature indicates there may be sexually dimorphic responses to poor prenatal sleep. Poor prenatal maternal sleep predicted presence of allergies, number of hospitalizations in the first year of life, and infant mortality more strongly in boys than girls (64,68). On the other hand, poor prenatal maternal sleep was linked to higher blood pressure and BMI, increased risk for obesity, hyperactivity, and behavior problems more so in girls than boys (65,67,69,76). These findings are consistent with the viability-vulnerability hypothesis, which supports that maternal signals affect male offspring via mortality, whereas female offspring show long-lasting developmental consequences (103). Poor prenatal maternal sleep may affect males and females differently, which would inform the creation of sex-specific interventions to improve offspring development.

Predictors of Prenatal Maternal Sleep

Poor prenatal maternal sleep may be a mechanism by which multilevel factors (i.e., social determinants of health) lead to poorer offspring health. Experiences of discrimination, redlining, shift work, access to care, and police brutality are related to sleep problems (104–107). Maternal life history, including both positive and negative experiences, and maternal mental health (45) are additional factors linked to prenatal maternal sleep (40,47). Future studies should continue to investigate factors that are associated with prenatal sleep as these may be key targets for future interventions.

CONCLUSION

Research on sleep health of pregnant individuals is understudied despite evidence that poor prenatal maternal sleep portends compromised offspring health and development. This review synthesized the existing literature on poor prenatal maternal sleep and its implications for offspring sleep and physical health, developmental outcomes, and brain structure and function. Pregnancy is a sensitive window in which interventions and policies can prevent sleep health disparities (108,109), and the present findings highlight the importance of addressing prenatal maternal sleep to improve health-related outcomes of pregnant individuals and their children.

Supplementary Material

Supplemental digital content is available for this article.

Abbreviations:

ADHD

attention-deficit/hyperactivity disorder

ASQ

Ages and Stages Questionnaire

BMI

body mass index

DOHaD

Developmental Origins of Health and Disease

EF

executive function

ERP

event-related potential

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

PSQI

Pittsburgh Sleep Quality Index

SDQ

Strengths & Difficulties Questionnaire