The Intervention Nurses Start Infants Growing on Healthy Trajectories (INSIGHT) responsive parenting (RP) intervention for first-time mothers improved firstborn infant sleep compared with controls. The goals of this analysis were to test intervention spillover effects on secondborn siblings and examine birth order differences in infant sleep.
Secondborns (n = 117) of INSIGHT mothers were enrolled in an observational cohort, SIBSIGHT. The Brief Infant Sleep Questionnaire was collected at 3, 16, and 52 weeks. Generalized linear mixed models assessed differences among secondborns by firstborn randomization, as well as birth order differences at 16 and 52 weeks.
The RP group secondborns slept 42 minutes longer at night (95% confidence interval [95% CI]: 19–64) and 53 minutes longer total (95% CI: 17–90) than control secondborns. RP secondborns were more likely to self-soothe to sleep (odds ratio [OR] = 2.0, 95% CI: 1.1–3.7) and less likely to be fed back to sleep after waking (OR = 0.5, 95% CI: 0.3–0.9) than secondborns of control mothers. RP secondborns were more likely to have a bedtime ≤8 pm at 3 (OR = 2.9, 95% CI: 1.1–7.7) and 16 weeks (OR = 4.7, 95% CI: 2.0–11.0). Few differences in sleep parenting practices were observed when comparing siblings within families. Secondborns slept 37 minutes longer than firstborns at 16 weeks (CI: 7–67, P = .03).
The INSIGHT RP intervention for first-time mothers had a spillover effect to secondborns, positively impacting sleep duration and behaviors. Intervening with first-time mothers benefits both firstborns and subsequent children.
The Intervention Nurses Start Infants Growing on Healthy Trajectories responsive parenting intervention for first-time mothers included education on consistent bedtime routines and nighttime parenting practices that promote self-soothing. Compared with controls, the intervention was associated with improved firstborn infant sleep behaviors and sleep duration.
Without further intervention, Intervention Nurses Start Infants Growing on Healthy Trajectories responsive parenting group families experienced a positive spillover effect on secondborn infant sleep health, such that secondborns had longer sleep duration and were more likely to self-soothe to sleep than secondborns of control group families.
Inadequate sleep during infancy is associated with adverse outcomes, including later risk for obesity.1–4 For optimal health, it is recommended that infants age 4 to 12 months regularly sleep 12 to 16 hours per 24 hours, including naps.5 Sleep-related parenting practices, such as consistent bedtime routines and developmentally-appropriate responses to night wakings, play a key role in helping infants develop healthy sleep patterns.6–8 Promoting an infant’s ability to self-soothe at night may reduce sleep disruptions for both infant and parent, and reduce reliance on nighttime feedings to soothe infants to sleep, which may prevent excess energy intake.9 Thus, interventions to promote development of good sleep habits in infancy may positively impact families in several ways.
The Intervention Nurses Start Infants Growing on Healthy Trajectories (INSIGHT) study is a parallel arm, randomized clinical trial comparing a responsive parenting (RP) intervention for primary prevention of obesity versus a safety control in firstborn infants.10 The RP intervention targeted parenting in several domains linked to obesity risk, including sleep. In addition to healthier weight outcomes,11,12 firstborn infants in the RP intervention group had longer nighttime sleep duration than controls and were less likely to be fed immediately before bed or back to sleep after night wakings. RP group infants were also more likely to have an 8 pm or earlier bedtime and were more likely to self-soothe to sleep.13
Because many families have more than one child,14 we developed SIBSIGHT, an observational study of secondborn children from families participating in INSIGHT, with the goal of determining whether the RP intervention delivered with firstborns has “spillover” benefits for secondborn children, as well as to examine birth order differences in obesity-related characteristics and parenting practices. We have previously demonstrated spillover effects of the RP intervention on infant body mass index (BMI),15 parent feeding practices,16 and dietary intake of secondborns.17 This analysis focuses on sleep-related outcomes, with objectives to: (1) determine whether the INSIGHT RP intervention delivered to mothers with their firstborn affected sleep outcomes of secondborn infants and (2) explore similarities within sibling pairs and birth order differences in infant sleep duration and behaviors. We hypothesized that mothers who received the RP intervention would continue to use this guidance with their secondborn, resulting in longer sleep duration and more optimal sleep-related behaviors in secondborns from RP group families versus controls.
Primiparous mother-newborn dyads were recruited in January 2012 through March 2014. Full details on recruitment, eligibility, randomization, and intervention have been previously described.10,13 Briefly, eligible mothers were ≥20 years old and English-speaking, and infants were singleton, ≥37 weeks gestation and ≥2500 g at birth. Enrolled participants (n = 279) were randomized to the INSIGHT RP intervention or a safety control intervention. Curricula were delivered by nurses at home visits at infant age 3 to 4, 16, 28, and 40 weeks and research center visits at 1 and 2 years, with additional messaging by phone at 18 and 30 months. The RP intervention included guidance on feeding, sleep, soothing or emotion regulation, and interactive play. Sleep messaging encouraged a consistent bedtime routine, an age-appropriate bedtime (7–8 pm), and to allow infants the opportunity to self-soothe to sleep at bedtime. Regarding night wakings, mothers were encouraged to allow infants an opportunity to self-soothe themselves back to sleep, and to use alternative soothing strategies (eg, offering a pacifier) rather than feeding infants who were not exhibiting signs of hunger. The control intervention was matched for intensity but focused on child safety; sleep-related messaging included sudden infant death syndrome (SIDS) prevention and crib safety. Families received up to $465 plus baby care gifts for participating in study visits and completing measures during the 3 year study.
INSIGHT-enrolled mothers from both the intervention and control groups who gave birth to a second child between June 2013 and March 2017 were invited to participate in the SIBSIGHT observational study (n = 138 screened). Secondborns were eligible (n = 122) if they were singleton, ≥36 weeks gestation and ≥2250 g at birth. All enrolled secondborns (n = 117) participated through the primary study endpoint at 52 weeks. Nurses conducted visits at 3, 16, 28, and 52 weeks for data collection, but no intervention content was delivered.10,16 Families received up to $250 for completing study visits and questionnaires. For both cohorts, mothers provided written consent for themselves and their children. The studies were approved by the Human Subjects Protection Office of the Penn State College of Medicine and registered at clinicaltrials.gov before enrollment of the first participant (NCT01167270). Participant flow for both studies is illustrated in Fig 1.
Demographic data were collected at enrollment. Mothers reported on infant sleep environment, bedtime routine activities, sleep duration, and response to nighttime awakenings using the Brief Infant Sleep Questionnaire (BISQ).6,18 Additional questions assessed specific behaviors recommended in the RP intervention. In INSIGHT, a short version of the BISQ was completed at infant age 2, 8, and 52 weeks, with full versions at 16 and 40 weeks. To match with secondborn data, only the 16 and 52 week firstborn data are used in the present analysis. In SIBSIGHT, a short version of the BISQ was completed at 3 weeks and the full BISQ was completed at 16 and 52 weeks. Surveys were completed electronically via Research Electronic Data Capture19 or on paper if needed.
Analyses were performed in SAS 9.4 (SAS Institute, Cary, NC). Statistical significance was defined as P < .05. Among secondborns, main effects of the firstborn study group and age of assessment (3, 16, and 52 weeks), and their interaction, were assessed. Categorical outcomes were analyzed using generalized linear mixed models with a spatial power covariance structure to account for repeated measures within subjects. Continuous outcome variables were analyzed using mixed linear models with a spatial power covariance matrix. Posthoc analyses of significant interactions were conducted by comparison of odds ratios or least-squared means.
Birth order differences were examined at 16 weeks, when the full BISQ was collected for both siblings, and on a limited number of variables at 52 weeks, given that a short version of the BISQ was used with firstborns at this time point. Only participants that completed surveys for both siblings at a given time point (n = 107 families at 16 weeks and 96 families at 52 weeks) were included in analysis. Fisher’s exact test was used to determine if firstborn sleep behaviors were associated with those in secondborns (ie, do mothers use the same sleep parenting practices with both children?), whereas generalized linear mixed models were used as described above to examine birth order differences in these variables (ie, are behaviors systematically different by birth order?). Pearson correlations were used to examine similarity in sleep duration between siblings. Birth order differences in sleep duration were assessed using linear mixed models, with repeated observations nested within sibling and siblings nested within family. Main effects of time and birth order, and their interaction, were tested, while controlling for study group assignment. Study group by birth order interactions were also examined.
Secondborns (n = 57 from RP families and n = 60 from control families) were delivered 2.5 ± 0.8 years after their firstborn siblings. Consistent with the full INSIGHT sample, mothers with an enrolled secondborn were predominantly non-Hispanic White, married, and college-educated. There were no significant study group differences in demographic or birth characteristics of secondborns (Table 1). Inclusion of demographic covariates (child sex, spacing between births, and sex constellation [ie, whether firstborn and secondborn were same or different sex]) in analyses did not change the results, so unadjusted analyses are presented below.
Intervention Spillover Efects on Secondborn Sleep Duration and Behaviors
Secondborns from RP families had significantly longer nighttime (7 pm to 7 am) and total 24 hour sleep duration than secondborns from control families at 3, 16, and 52 weeks (Fig 2). Nighttime sleep increased with age, whereas daytime (7 am to 7 pm) and total sleep decreased with age (all P < .001), but there were no significant study group by age interactions, suggesting that the effects of the firstborn intervention group were consistent across the 3 time points. On average across all time points, secondborns from RP families slept 41.6 (95% CI: 18.9–64.2) minutes longer at night and 53.3 (16.8–89.8) minutes longer over 24 hours than secondborns from control families. There was no group difference at any time point in secondborn daytime sleep duration.
Secondborns from RP families had greater odds of being read to during their bedtime routine than control secondborns (Table 2); this was consistent across 16 and 52 weeks (OR = 2.6 [1.3–5.4], P = .01). Following intervention recommendations, RP group mothers were less likely to rock their secondborn before bed at 52 weeks (OR = 0.4 [0.2–0.8], P = .01), with no difference at 16 weeks. Though uncommon in both groups, fewer RP than control secondborns watched television during their bedtime routine at both 16 (0 vs 4%) and 52 (2 vs 13%) weeks. There were no study group differences or group by time interactions in the percent of mothers who reported, including bathing, music, or feeding in their secondborn’s bedtime routine.
There was a significant study group by time interaction for the odds of secondborns meeting bedtime recommendations (Table 2). Secondborns from RP families were more likely than controls to have a bedtime of 8 pm or earlier at 3 (OR = 2.9 [1.1–7.7], P = .04) and 16 (OR = 4.7 [2.0–11.0], P < .001) weeks, with no difference at 52 weeks. At 3, 16, and 52 weeks, secondborns from RP group families had higher odds of self-soothing to sleep (OR = 2.0 [1.1–3.7], P = .02), and lower odds of falling asleep while being held (OR = 0.5 [0.3–0.9], P = .02). RP group mothers were also more likely than controls to report that their secondborn took 15 minutes or less to fall asleep (OR = 2.2 [1.1–4.3], P = .02). At 16 weeks, a greater percentage of RP group secondborns fell asleep while swaddled (43 vs 22%, χ-Sq P = .02). There were no group differences or group by time interactions in the percentage of secondborns who fell asleep with a pacifier or while listening to white noise.
Consistent with the RP intervention recommendations, RP group mothers were less likely than control mothers to report holding or rocking (OR = 0.4 [0.2–0.6], P <. 001) or feeding (OR = 0.5 [0.3–0.8], P = .01) their secondborn to sleep following a night waking (Table 2); these differences were consistent across all 3 time points. There was a significant group by time interaction for changing the baby’s diaper in response to night waking, but posthoc comparisons between study groups at each time point were not statistically significant. There were no group differences or group by time interactions in the number of mothers who reported frequent night wakings and night feedings or giving their secondborn a few minutes to fall back asleep, rubbing or patting, or giving a pacifier after waking at night (Table 2).
Birth Order Differences in Sleep Duration and Behaviors
At 16 weeks, there were modest correlations between firstborn and secondborn daytime (r = .35, P < .001) and total 24-hour (r = .27, P = .005), but not nighttime (r = .15, P = .14) sleep duration. At 52 weeks, similar correlations were seen between siblings for nighttime (r = .30, P = .004), daytime (r = .30, P = .004), and total (r = 0.35, P <.001) sleep duration. Birth order effects on sleep duration varied between 16 and 52 weeks and nighttime versus daytime sleep (Fig 3). For nighttime sleep, there was a significant time by birth order interaction (P = .02) such that firstborns slept longer than secondborns at 52 weeks only. There was also a significant time by birth order interaction for daytime (P = .04) and total sleep (P = .002) such that secondborns slept longer than firstborns at 16 weeks only.
Similarities and differences in sleep behaviors and parenting practices between firstborns and secondborns at 16 weeks are presented in Table 3. For each question, around 60% to 70% of mothers reported using the same behaviors with the secondborn as they did with their firstborn. The majority of bedtime routine components did not differ by birth order; however, secondborn bedtime routines were less likely to include reading (OR = 0.4, 95% CI: 0.2–0.7, P < .001), baths (OR = 0.6, [0.4–0.9], P = .03), and watching television (OR = 0.2, [0.04–0.9], P = .04) than those of firstborns. Compared with firstborns, secondborns were less likely to sleep in their own room (OR = 0.6, [0.4–0.8)] P = .004). There were no birth order differences in meeting the ≤8 pm bedtime recommendation, falling asleep while swaddled, with a pacifier, with white noise, while being held, or self-soothing to sleep, or in the odds of taking 15 minutes or less to fall asleep. Regarding sleep safety, most firstborns and secondborns were put to sleep in the recommended supine position, but secondborns were less likely than firstborns to have objects such as blankets, pillows, or stuffed animals in their crib (OR = 0.6 [0.4–0.9], P = .02). There was no difference by birth order in odds of frequent night wakings or feedings, or in use of any parenting strategies in response to night waking. There were also no significant birth order by study group interactions for any outcomes.
Mothers who were randomized to the INSIGHT RP intervention with their firstborn reported longer infant sleep duration and more frequent use of responsive sleep parenting practices with their secondborn, compared with mothers who were randomized to the control group with their firstborn. These study group differences are similar to those we previously reported for firstborns,13 suggesting an intervention spillover effect on infant sleep to secondborns despite mothers receiving no additional intervention for these younger siblings. Secondborns from RP families slept significantly longer than those from control families, and differences were consistent across measurement time points during the first year. Similar to findings with firstborns,13 RP group mothers were more likely than control mothers to use an age-appropriate bedtime for their secondborn, to read to their second child as part of a bedtime routine, and to allow their second child to self-soothe themselves to sleep and were less likely to feed their second child back to sleep after waking. Paralleling our previous findings on infant BMI,15 feeding,16 and diet,17 these data suggest that intervening with first-time mothers has potential to benefit not only firstborns but subsequent children as well.
Sleep health is influenced by both genetic and environmental factors. In twin studies of young children, shared environmental influences (eg, home environment) have the greatest contribution to sleep duration and quality, explaining 55% to 83% of variance, with additive genetic components contributing 5% to 40% and nonshared environmental factors contributing 1% to 17%.20–22 For the majority of sleep-related behaviors surveyed, around 70% of mothers reported concordant answers for both siblings. These data suggest that, regardless of intervention, early life sleep parenting practices are fairly consistent across firstborn and secondborn siblings, likely contributing to the importance of shared environment. Somewhat lower concordance was observed for items that may be more reflective of infant characteristics, such as time taken to fall asleep and frequent night wakings, compared with those that are parent behaviors. Biological factors and other individual differences (eg, temperament) may contribute to variability in sleep between siblings, warranting future investigation.
Although we observed modest correlations in infant sleep duration between siblings, secondborns had longer daytime and total sleep than firstborns at 16 weeks, but shorter nighttime sleep than firstborns at 52 weeks. Consistent with our 16 week findings, a study of 4 to 9 week old infants found that those who were firstborns had shorter nap and 24 hour total sleep duration than subsequent-born infants, with no difference in nighttime sleep duration.23 Other studies have found no association between maternal parity or number of siblings and sleep duration during infancy and early childhood.3,24 We also observed some birth order differences in sleep-related parenting practices. The bedtime routines of secondborn children were less likely to include reading books and baths than those of their firstborn sibling. With the arrival of a second child, mothers experience increased time demands resulting in “resource dilution”25 and less 1-on-1 time with each individual child26,27 and often need to develop new routines to establish healthy sleep patterns for both children.28 Secondborns were also less likely than firstborns to watch television as part of their bedtime routine. Some previous research has found greater use of screen media among infants and young children without siblings,29,30 though others have found no association.31–33 Secondborns were less likely than firstborns to sleep in their own room rather than in their parents’ room, another finding consistent with previous research.34 However, as the number of children increase, parents may have fewer options for child sleep locations and room-sharing with parents may be preferable to having infants sleep in the same room as older siblings.
A strength of this study is that we were able to compare sleep behaviors in infancy among siblings within the same family. However, these families were mostly non-Hispanic White with few lower-income households, limiting the generalizability of our findings. The SIBSIGHT sample is smaller than the original INSIGHT randomized trial. A larger sample may have yielded more power to detect study group differences in secondborns. Mothers were the primary intervention target for INSIGHT, though fathers and other caregivers were encouraged to attend study visits and review materials. Future interventions may be strengthened by more directly targeting all caregivers that contribute to sleep parenting. We did not collect concurrent information on firstborn sleep at the time of secondborn data collection, limiting our ability to assess how firstborn daily routines influence secondborn sleep. Additionally, sleep outcomes were reported by mothers via questionnaire. Though sleep duration and night wakings as measured by the BISQ have been shown to correlate with actigraphic measures,18 a more objective measure of sleep, such as actigraphy, would strengthen the results.
In summary, this analysis demonstrates that an RP intervention, including guidance on infant sleep, delivered to first-time mothers has positive spillover effects for secondborn infant sleep. Although there were some differences by birth order in sleep outcomes, mothers’ sleep parenting of their first and secondborn child was generally more alike than different, further suggesting that interventions targeting first-time parents may have sustained benefits for future children. Clinicians should help first-time parents establish consistent bedtimes, bedtime routines, and appropriate responses to night wakings early in infancy to help firstborn and subsequent children develop healthy sleep routines. Although home-delivered parenting interventions are resource intensive, intervening on first-time parents may be a cost-effective way to reach multiple children within a family. Scalability of future interventions may be improved by incorporating the INSIGHT RP guidance into existing primary care and home visiting programs or delivering via telehealth.
We thank Leann Birch, PhD, Jodi Mindell, PhD, Jessica Beiler, MPH, Jennifer Stokes, RN, Amy Shelly, LPN, Patricia Carper, RN, and Lindsey Hess, MS for their contributions to this project.
Dr Hohman analyzed and interpreted the data and drafted the initial manuscript; Drs Savage and Anzman-Frasca wrote the intervention curriculum, contributed to study design, interpretation of the data, and critical revision of the manuscript; Dr Loken and Ms Marini contributed to study design, interpretation of the data, and critical revision of the manuscript; Dr Buxton contributed to interpretation of the data and critical revision of the manuscript; Dr Paul led the conception and design of the study, and contributed to interpretation of the data and critical revision of the manuscript; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
This trial has been registered at clinicaltrials.gov, (identifier NCT01167270).
FUNDING: Funded by the National Institute for Health grants R01DK088244 and R01DK099364. Research Electronic Data Capture support was received from the Penn State Clinical and Translational Sciences Institute, NIH UL1 TR002014. Funded by the National Institutes of Health (NIH).
CONFLICT OF INTEREST DISCLOSURES: The authors have indicated they have no conflicts of interest to disclose. Outside of the current work, O.M.B. received subcontract grants to Pennsylvania State University from Proactive Life (formerly Mobile Sleep Technologies) doing business as SleepSpace (National Science Foundation grant 1622766 and National Institute for Health and National Institute on Aging Small Business Innovation Research Program R43AG056250, R44 AG056250), honoraria or travel support for lectures from Boston University, Boston College, Tufts School of Dental Medicine, Harvard Chan School of Public Health, New York University, and Allstate, consulting fees from SleepNumber, and an honorarium for his role as the Editor-in-Chief of Sleep Health (sleephealthjournal.org).
Brief Infant Sleep Questionnaire
Intervention Nurses Start Infants Growing on Healthy Trajectories