Effect of a behavioral-educational sleep intervention for first-time mothers and their infants: pilot of a controlled trial
ABSTRACT
Study Objectives:
This study tested the acceptability and efficacy of a perinatally delivered behavioral–educational sleep intervention.
Methods:
Participants were 40 primiparous women assigned in late pregnancy to either an intervention (n = 20) or control (n = 20) group. The sleep intervention group (SIG) received prenatal anticipatory education and guidance regarding their own and their infant’s sleep during the first 3 months postpartum. This was reinforced during phone calls within the first 6 weeks postpartum. The control group (CG) received brief sleep hygiene information at a prenatal session, followed by 2 phone calls during the same period. Mother–infant pairs wore actigraphs for 48 hours at 6 and 12 weeks postpartum, and mothers kept sleep diaries. Questionnaires completed in late pregnancy and 6 and 12 weeks postpartum related to sleep, newborn care, and mood. The main outcome measures included maternal sleep quantity, efficiency, and self-reported quality and infant sleep duration and consolidation.
Results:
Mothers reported high acceptability of the study processes. Sleep duration and quality increased for mothers and infants across time in both groups, with a significantly greater increase in nocturnal sleep duration for mothers in the SIG.
Conclusions:
Prenatal sleep guidance and postnatal follow-up seems to enhance nocturnal sleep of mothers, change their perceptions of their own sleep, and increase confidence in managing their infant’s sleep. Follow-up at later intervals and replication with larger, more diverse samples may reveal further differences.
Citation:
Sweeney BM, Signal TL, Babbage DR. Effect of a behavioral-educational sleep intervention for first-time mothers and their infants: pilot of a controlled trial. J Clin Sleep Med. 2020;16(8):1265–1274.
BRIEF SUMMARY
Current Knowledge/Study Rationale: Sleep is a significant aspect of early infancy and can be stressful for the whole family. Sleep is linked to maternal mental health, with anxious or depressed mothers reporting difficulties falling asleep even when their infant is settled. Little research-based behavioral sleep education is given to either health practitioners or parents who may have limited understanding of normal infant sleep development or practices that could support family sleep.
Study Impact: Prenatal sleep guidance and postnatal follow-up appears to enhance nocturnal sleep of mothers, change their perceptions of their own sleep, and increase confidence in managing their infant’s sleep. A positive transition to parenting may be enhanced by the provision of evidence-based behavioral–educational sleep interventions.
INTRODUCTION
A fundamental part of the newborn’s transition from “womb to world” is development of biological rhythms and sleep–wake patterns. Initially, this can seem disorganized and exhausting in the early postpartum weeks.1 Early establishment of the infant’s sleep–wake patterns occurs parallel to the mother’s recovery from pregnancy, parturition, establishment of feeding, and the critical psychological transition to becoming a mother. Perinatal physical recovery and psychological adaptation are negatively impacted when sleep is poor.2
Searching the phrase “infant sleep” returns tens of millions of website, blog, social media, and mobile app results. This and the proliferation of infant sleep consultants and coaches3 are testament to the vested interest new parents have in their own and their infant’s sleep. Narratives on coping with the expected sleep disruption that accompanies a newborn are often emotive. Some parents seek unambiguous, didactic advice on “how to,” and others prefer a flexible self- or infant-guided approach, but all seek to be supported in their parenting role.
Sleep is a significant aspect of early infancy and can be stressful for the whole family. Sleep is linked to maternal mental health, with anxious or depressed mothers reporting difficulties falling asleep even when their infant is settled.4 Little training is given to health practitioners,3 and less research-based education is provided for novice or experienced parents.5
Numerous interventions have targeted sleep in infants older than 6 months, some also focusing on maternal well-being.6–9 Few studies have investigated the effects of sleep interventions in the first weeks postpartum, either maternal10 or infant, and fewer have addressed dyad sleep.11 There is ongoing debate about the appropriateness of extinction-based methods in early months while feeding establishes,12 and there is rapid maturation in circadian and sleep physiology.13,14 Some parents find extinction-based methods unacceptable because of protracted periods of infant crying,15,16 but this does not preclude education being offered about normal sleep development and guidance on behaviors that may help or hinder reaching family sleep goals. Timing can be as critical as intervention content. Prenatally, women are often open to new health information and behavior change17,18 and seek strategies for coping with a newborn. Parents may particularly benefit from education and guidance before they are faced with the realities of newborn needs, their own recovery from birth, and transition to parenting.
In this study, a behavioral–educational sleep intervention was developed: the Parent Information for Parent and Infant Sleep (PIPIS) intervention. It was formatted to be incorporated into existing prenatal programs or offered in addition to usual education programs or prenatal care. The PIPIS intervention comprised prenatal education and postnatal support and was assessed for both efficacy and acceptability.
METHODS
Design and participants
Participants were assigned to a control or intervention group based on recruitment site. The New Zealand Health and Disability Ethics Committee (CEN 09/09/070) approved the study. Pregnant women attending community childbirth education classes in Wellington (New Zealand) were eligible to enroll and receive gift vouchers on completion of each study stage, to a total of NZ$110. Inclusion criteria included the following: primaparous; ≥16 years of age; singleton pregnancy; and planning to provide full-time infant care for at least 12 postnatal weeks. Women were excluded if they or their partner had experienced a stillbirth or perinatal death at ≥20 weeks of gestational age; they had a partner with children from another relationship; they had current uncontrolled chronic illness; they used prescription medications that affected sleep; and they or their partner had a diagnosed sleep disorder.
The researcher introduced the study to women in the last trimester of pregnancy at a community childbirth education class (Figure 1). Recruitment took place from September 2010 to February 2011. Women were recruited through a single-provider organization offering classes in two demographically similar, but geographically distinct, areas. Recruitment occurred across several 7-week classes. Sites alternated as either an intervention or control condition for each round of recruitment. In total, four rounds of recruitment were necessary to obtain the required sample size. Every effort was made to conceal group assignment; full blinding was not possible, because a requirement from the guiding ethics agency was that the possibility of assignment to 1 of 2 groups be specified, stating each would receive different levels of sleep information and contact.
Sleep intervention
The intervention comprised a 2-hour, small group, prenatal education session with follow-up telephone support. In all, 5 groups were facilitated, ranging in size from 2 to 6 participants. Women received a session content booklet and relaxation audio recording. Topics included the following: normal adult sleep stages, quantity, quality, and timing; consequences of poor sleep; common changes to perinatal sleep; normal infant sleep development; and strategies for optimizing sleep in the dyad. A transactional model of infant sleep, based on Sadeh et al,19,20 helped frame expectations about infant sleep in their unique family context, and establishing feeding was emphasized as foundational to setting the scene for sleep. Cognitive and behavioral strategies to optimize maternal and infant sleep were included, but parents were not directed to adhere to any particular strategy. Mothers in SIG were referred back to these resources during weekly follow-up calls initiated over 5 weeks, starting after postnatal day 7. It was not possible to complete every scheduled call because women were napping, out, busy with baby care, and so on. All women took part in at least 2 calls, and 75% took part in 4 or more weekly calls. Participants were encouraged to voice questions or concerns about their or their infant’s sleep.
Control condition
A 1-hour, small group (n = 21) information session covered general sleep hygiene tips, explanation of objective sleep monitoring, and study administration. In all, 6 groups were facilitated, ranging in size from 2 to 7 participants. Questions about the study process only were encouraged. No ongoing telephone support was offered beyond brief calls made at postnatal weeks 2 and 4 by a researcher who did not present to the group, who encouraged continued study involvement.
Partners in both groups were encouraged, but not required, to attend the group sessions.
Outcome measures
Between 35 and 37 weeks of gestation, all women completed a Sleep and Health During Pregnancy Questionnaire, developed for an associated study.21 A postnatal version of the Sleep and Health questionnaire was completed at 12 weeks postpartum. Questionnaires included sleep items and asked about planned and actual infant feeding and sleep location and whether mothers perceived infant sleep to be a problem. The Edinburgh Postnatal Depression Scale (EPDS)22 was used to screen for symptoms of perinatal depression and anxiety. The EPDS is one of the most widely used self-report screening instruments for symptoms of mental health difficulties (depression and anxiety) in the perinatal period. The EPDS has also been validated for use in the prenatal period.23
Self-reported sleep quality
A single question “[Before this pregnancy] or [In the last week] how often did you get a good night’s sleep” (based on the National Sleep Foundation)24 was used as a measure of self-reported sleep quality, measured retrospectively for prepregnancy sleep, currently at 35–37 weeks of gestation, and at 12 weeks postpartum. Response options ranged from 0 (no nights) to 7 (every night).
Objective sleep measurement
All women and infants completed 48 hours of continuous actigraphic monitoring at 6 and 12 weeks postpartum using the Actiwatch-64 device (Mini-Mitter, Bend, OR). Published actigraphy guidelines informed data collection and analysis (including those of the American Academy of Sleep Medicine,25 Sadeh,26 Meltzer et al,27 and Ancoli-Israel28). Mothers wore the actigraph on their nondominant wrist and infants around the midcalf. Infant devices were contained in a bespoke, soft, size-adjustable, stretch band. Mothers received face-to-face instruction on the actigraph’s capability, safety, and how to wear and use the device, including tips on infant clothing to allow ease of access to the device, such as to press the event button. Mothers were also given a handout detailing the same information and encouraged to telephone the researcher if they had any questions or concerns. Wherever possible, the monitoring days were scheduled to avoid weekends, when sleep patterns may differ from weekdays because of reduced work commitments for partners. During a telephone call to schedule an actigraphy setup home visit, mothers were asked not to schedule immunizations in the 48 hours before the start of monitoring. Common adverse reactions to pediatric immunization include fussing, crying, fever, needle site inflammation, and sleepiness,29 and the presence of these may influence sleep during the subsequent 24 hours. These reactions may also influence maternal responses to questionnaire items about infants’ sleep habits.
Manufacturer supplied software (Actiware, Version 5) was used to configure devices (1-minute epochs) and score rest/activity data. Previously, the medium wake sensitivity threshold (40 activity counts/min) has been found to achieve the highest agreement (κ) and closest estimate of sleep epochs defined by polysomnography in adults.30 Infants have been observed to be more active during sleep,31 and for this, reason a high wake threshold sensitivity (80 activity counts/min) was used to distinguish infant sleep from waking activity.32,33 Mothers pressed an event button on their device and their infant’s device to coincide with start and end of rest opportunities.
Sleep diaries (actigraphy logs) were used to corroborate actigraphy records. These captured additional information about events of note including whether sleep periods included external motion such as during car journeys, times when the device was removed (such as for bathing), and feeding periods. Mothers recorded nighttime sleep locations on the infant diary. Double-agreement scoring was performed on 20% of mother and infant actigraphy files (randomly selected), and a >80% agreement rate34 was met.
Actigraphy has been reported as valid for use with young infants.25 Challenges in using actigraphy in very young infants include artifact from external motion35 and high levels of activity and arousals during sleep that are associated with normal early infant neurologic development.36 The present study was not immune to these challenges, and after data collection and initial visual screening of actograms, reliability of the classification of infant sleep by Actiware became a concern. For these reasons, the decision was made to use rest duration (time in bed [TIB]) as the main infant analysis parameter. Combined use of the actigraph, event marker, and log information was aimed at enhancing the accuracy of the actigraphy data alone.
Objective sleep variables
Night was defined as 2100 to 0900 hours, following the work of Stremler et al.11 Maternal sleep measures included TIB, total sleep time (TST), total nocturnal TST, longest nocturnal sleep episode, and nocturnal sleep efficiency. Infant sleep measures included number of night awakenings, total nocturnal TIB (Noct-TIB), and longest Noct-TIB period.
Pilot study feedback
A final questionnaire sought feedback on women’s experience of participating in the study and a postintervention measure of maternal self-efficacy in relation to infant sleep. Once data collection was complete, each woman was sent a letter advising that the study had been a trial and her group assignment. Participants were sent the materials that the alternate group received and given the opportunity to experience the corresponding session. One participant from each condition chose this.
Sample size
Although the PIPIS study was run as a pilot study, power analyses were conducted to determine sample size. Previous pilot work, similar in design to PIPIS, was successfully completed with a total sample size of 30 postnatal women (15 controls and 15 intervention women). This intervention achieved a between-group difference in nocturnal TST of 57 minutes, with a large effect size (d = 2.94 at α = .05).11 We also tested a more conservative difference in nocturnal TST of 30 minutes, using the same sample size, and a large effect size remained (d = 1.55 at α = .05). A target sample of 20 participants per group was therefore chosen to allow for potential dropout.
Statistical analysis
Data were screened using both statistical and graphical methods according to the checklist for screening data described by Tabachnick and Fidell.37 Accuracy of data input was assessed by inspection of univariate descriptive statistics to detect out-of-range values, plausible means and standard deviations, and univariate outliers. All databases were inspected to gauge the amount and pattern of missing data. The amount of missing data was minimal (<2%). Descriptive statistics were generated using SPSS (Version 19; IBM Corp, Armonk, NY).
Infant feeding mode and sleep location
Breastfeeding status was originally considered for inclusion in analyses because frequent sleep disturbance to feed their infants (≥3 nights per week) was noted by 80% of mothers in the CG and 75% of mothers in the SIG at 12 weeks postpartum. On assessment, there was little between-groups difference in breastfeeding status, and the number of women who were formula feeding were so few it was not included in any final models. Based on preliminary within- and between-group differences, infant sleep location was included as a fixed effect in all mixed models where infant sleep location applied.
Generalized linear mixed-effects models were used to examine changes in survey and actigraphy data across time using SAS (Version 9.3; SAS, Inc., Cary, NC). Maternal and infant objective sleep models included time, group, and location as factors, as well as interactions for time × group and group × location. A compound symmetry covariance structure, with Kenward-Roger correction,38 was applied to all analyses. For all tests, the type I error rate was set at .05.
RESULTS
In total, 41 women completed all stages of the PIPIS study. All were living together with a spousal partner. Most identified as New Zealand European (n = 33). Groups did not differ on socioeconomic deprivation deciles using the New Zealand Deprivation Index, an area-based measure of relative deprivation.39 Participants represented all but the most deprived decile. One dyad was excluded from analysis because the infant was born with a genetic condition known to impact sleep. Characteristics of participants at baseline (35–37 weeks of gestation) are shown in Table 1.
| Characteristic | SIG (n = 20) | CG (n = 20) |
|---|---|---|
| Maternal age, mean (SD) | 33.7 (2.9) | 32.9 (2.8) |
| Participating in work for pay (n, %) | 17 (85) | 19 (95) |
| Prenatal parenting plans | ||
| Planning to breastfeed (n, %) | 14 (70) | 12 (60) |
| Planning to room share (n, %) | 13 (65) | 15 (75) |
| Planning to bed share (n, %) | 0 (0) | 0 (0) |
| EPDS > 12 (n, %) | 2 (10) | 0 (0) |
| Sleep quality | ||
| Good night’s sleep per week, mean (SD) | 2.9 (1.9) | 3.5 (1.7) |
| Good night’s sleep per week | ||
| Retrospective prepregnancy report, mean (SD) | 5.9 (1.0) | 6.1 (.8) |
| Delivery | ||
| Gestational weeks at birth, mean (SD) | 39.3 (1.7) | 39.8 (1.2) |
| Induction of labor (n, %) | 8 (40) | 10 (50) |
| Vaginal birth (n, %) | 14 (70) | 16 (80) |
| Instrumental delivery (forceps, ventouse) (n, %) | 6 (30) | 5 (25) |
| Caesarean section (n, %) | 6 (30) | 4 (20) |
| Female infant (n, %) | 12 (60) | 6 (30) |
Maternal objective sleep duration
A total of 274 and 213 episodes of rest/sleep actigraphy were evaluated at 6 and 12 weeks postpartum, respectively. There were no missing actigraphy values for variables of interest at either time point. There were no significant effects for TST in 24 hours across the postpartum period within or between groups. For the total nocturnal TST, the time by group interaction was significant. Post hoc analysis showed a significant increase for SIG only. Nocturnal sleep in this group increased by 47 minutes over the second 6 weeks postpartum [t(36.55) = −4.30, P < .001; Figure 2], with no significant increase for CG. Between-group differences in total nocturnal TST at 6 and 12 weeks were not significant in post hoc analyses. The longest nocturnal sleep episode increased significantly for both groups from 6 to 12 weeks by an average of 48 minutes, but there were no between-group differences.
Error bars represent standard error.
Maternal sleep quality
Objective sleep episodes and efficiency
Both groups experienced a decrease in the number of sleep episodes in 24 hours and the number of sleep episodes at night between 6 and 12 weeks postpartum. Within-group sleep efficiency was relatively stable across the study (Table 2). There was a statistically significant main effect of group for sleep efficiency in 24 hours so that CG efficiency was higher (85.3%) compared with mothers in the SIG [82.9%; t(38.3) = 2.07; P = .045]. Although this difference is statistically significant, a 2.4% difference in sleep efficiency may represent little clinical or practical effect. No significant differences were found for nocturnal sleep efficiency.
| Characteristic | SIG (n = 20) | CG (n = 20) | ||
|---|---|---|---|---|
| 6 wk | 12 wk | 6 wk | 12 wk | |
| Mother | ||||
| EPDS > 12, n | 0 | 0 | 1 | 2 |
| Sleep quality | ||||
| Good night’s sleep per week, mean (SD) | 3.9 (2.1) | 2.9 (2.6) | ||
| Objective sleep | ||||
| TST 24 hours (min), mean (SD) | 392.6 (55.9) | 419.8 (58.1) | 412.9 (55.6) | 412.9 (42.8) |
| Nocturnal TST (min), mean (SD) | 368.9 (54.0) | 415.7 (58.2) | 388.1 (56.9) | 402.1 (36.8) |
| Longest nocturnal sleep period (min), mean (SD) | 214.4 (61.4) | 270.3 (107.8) | 198.8 (44.9) | 250.7 (64.9) |
| Sleep episodes 24 hours, mean (range) | 3.3 (1.0–7.3) | 2.6 (1.0–5.0) | 3.5 (2.0–6.0) | 2.7 (1.0–6.0) |
| Nocturnal sleep episodes, mean (range) | 3.0 (2.0–4.0) | 2.5 (1.5–4.5) | 3.0 (2.0–4.5) | 2.4 (1.0–4.0) |
| Sleep efficiency 24 hours (%) | 83.6 (4.4) | 82.4 (5.7) | 84.4 (4.9) | 85.5 (2.6) |
| Nocturnal sleep efficiency (%) | 82.6 (5.5) | 81.8 (6.1) | 83.0 (7.4) | 84.7 (2.9) |
| Infant | ||||
| Fully breastfeeding,a n (%) | 19 (95) | 18 (90) | 18 (90) | 17 (85) |
| Parental room for night sleep, n (%) | 13 (65) | 12 (60) | 12 (60) | 7 (35) |
| Being held for sleep any part of night, n (%) | 3 (15) | 1 (5) | 11 (55) | 2 (10) |
| Bed sharing with parent at any time during night, n (%) | 4 | 2 | 6b | 1 |
| Infant sleep considered a problem, n (%) | 14 (70) | 11 (55) | 11 (55) | 10 (50) |
| Objective sleep | ||||
| Total rest time 24 hours (min), mean (SD) | 814.6 (111.0) | 859.8 (119.7) | 879.8 (88.0) | 890.4 (87.1) |
| Total nocturnal rest (min), mean (SD) | 524.2 (54.1) | 582.6 (55.5) | 549.8 (47.5) | 587.3 (47.8) |
| Longest nocturnal rest period (min), mean (SD) | 273.5 (71.8) | 341.6 (110.7) | 268.6 (57.1) | 354.8 (97.6) |
| Longest daytime rest period (min), mean (SD) | 126.1 (47.0) | 103.2 (47.2) | 137.2 (61.6) | 119.3 (49.9) |
| Rest episodes 24 hours, median (range) | 7.2 (4.9–12.2) | 7.3 (4.9–10) | 6.8 (4–10) | 6.8 (4–9.4) |
| Noctural rest episodes, median (range) | 4.0 (2–5) | 3.0 (2–4.5) | 3.5 (3–5) | 3.0 (2–4.5) |
Self-reported sleep quality
On average, good night’s sleep (GNS) in both groups decreased across the prenatal period and did not return to the prepregnancy baseline by 12 weeks postpartum (Table 2). A main effect of time was found for GNS [F(2,78) = 39.07; P < .001]. Participants reported getting, on average, 6 GNS per week before pregnancy, which decreased to 3.15 GNS at 35–37 weeks of gestation [t(78) = 7.95; P < .001]. Between late pregnancy and 12 weeks postpartum, there was no change in GNS [t(78) = −.63; P = .532]. There was no significant interaction effect of time × group for GNS [F(2,76) = 2.53; P = .09]; however, there was a trend for mothers in the SIG to have a greater number of GNS than mothers in the CG at 12 weeks postpartum (Figure 3).
Error bars represent standard error.
Infant objective rest duration
TIB in 24 hours did not increase significantly over time for infants, but there was a main effect of location. On average, total TIB in 24 hours was 40 minutes longer when infant night sleep was in the parental bedroom [F(1,73) = 4.04; P = .048].
The total Noct-TIB model produced a main effect of time. On average, there was an increase in Noct-TIB from 538 minutes per night at 6 weeks to 588 minutes per night at 12 weeks postpartum (+50 minutes per night). The interaction of group × location was also significant in this model [F(1,67) = 6.16; P = .02]. Post hoc analyses (Figure 4) showed that Noct-TIB was significantly greater for infants in the CG who slept in the parental bedroom (580 minutes) compared with infants in the SIG who slept in the parental bedroom [542 minutes; t(54.97) = 2.21; P = .031]. There was no significant difference in Noct-TIB between CG and SIG infants sleeping out of the parental bedroom [t(60.33) = −.76; P = .452].
Error bars represent standard error.
The model for the longest Noct-TIB produced a main effect for time; on average, infants’ longest period of Noct-TIB increased from 272 minutes at 6 weeks postpartum to 348 minutes at 12 weeks postpartum (+76 minutes).
Infant sleep location, nighttime awakenings, and sleep as a problem
By 12 weeks postpartum, 60% of infants in the SIG and 35% of infants in the CG remained in the parental bedroom for night sleep (Table 2). At 12 weeks postpartum, based on objective data, all infants were still having at least 2 sleep episodes between 2100 and 0900 hours, meaning all infants were waking at least once per night. At least one half of parents reported their infant’s sleep to be a problem across the study (Table 2).
Feedback and feasibility
Study evaluations were returned by 93% of participants, most of whom rated questionnaire completion and telephone contact easy (Figure 5). Participants found the actigraphic monitoring to be the most difficult, especially the practicalities of pressing the device event marker button when infants had already settled for sleep or remembering to do so. Mothers commented this was easier at 12 weeks than 6 weeks. Despite minimal prenatal sleep information provided to mothers in the CG mothers, more than one half (55%) reported they would definitely recommend this to others, and 40% said maybe. Only mothers in the SIG received the booklet (90% who would definitely recommend), support calls (71% reporting very useful), and sleep intervention (75% reporting very helpful for mother and infant sleep).
Participants were asked to rate their confidence on a 10-point self-reported rating of confidence scale (1 = never confident, 10 = always confident) across 6 sleep-related behaviors of infants (eg, able to recognize tired cues of the infant, putting baby to bed awake and sleep ready). Mothers in the SIG were more confident at being able to recognize the tired cues of their infant [t(33) = −2.247; P = .03], and a trend was seen for mothers in the SIG to report higher mean confidences scores on all other items.
DISCUSSION
In this pilot study, we found good feasibility and acceptability for the PIPIS intervention. To our knowledge, no other intervention has offered a prenatal behavioral–educational session with postnatal focus and follow-up on the mother–infant dyad. Most mothers in both groups reported satisfaction with the program, despite the limited support or education given to the CG. This suggests a need for any assistance new parents can get, possibly not just with sleep. Women appreciated the opportunity to participate and the contact they received, albeit limited for the CG. Comments from the CG indicated they were seeking the very information being offered to the SIG: “Own sleep dependent on babies [sic] sleep so more tips on baby’s sleep [would be] useful. Tired signs. Indication of how much sleep your babies generally need.”
Objectively measured maternal total sleep in 24 hours did not significantly increase across the study nor did nocturnal sleep for CG women. However, women who completed the behavioral–educational sleep intervention experienced a significant increase in nocturnal sleep from 6 to 12 weeks postpartum. Given that both groups experienced the same number of sleep episodes and infant sleep was not significantly different by group, one explanation for this difference in increase may be that SIG women were able to initiate sleep or return to sleep more quickly after night disturbances. This may also be reflective of SIG women reporting greater confidence in managing their infant’s sleep, although further work is needed to conclude such a relationship. Preliminary findings describing a positive relationship between objectively measured maternal sleep continuity and sensitivity of mothers to their infants during play at 18 weeks postpartum40 highlight the importance of promoting optimal sleep to support parent-infant relationships.
Sleep consolidation over time was evident as shown by the longest nocturnal sleep episode increasing significantly for both groups of women from 6 to 12 weeks postpartum, with no significant difference between the groups at 12 weeks. Also of interest was the trend for mothers in the SIG to report more GNS at 12 weeks postpartum than mothers in the CG (3.9 vs 2.9 nights/wk). Although this difference did not reach statistical significance in the current study, when combined with the higher confidence ratings for managing infant sleep, these findings may indicate a more positive maternal perception of sleep for the dyads in the SIG. This finding shows a reverse in the downward trajectory for the number of GNS that mothers in the SIG reported, whereas this downward trend continues for mothers in the CG at 12 weeks. The differences in numbers of mothers reporting any sleep problem in their infants was not significant at either time point; however, the fact that half of both groups report their infant’s sleep to be a problem at 3 months suggests ongoing support may be appropriate for parents across the first postnatal year. Future studies should include follow-up at least until 12 months to see if these trends persist.
By 12 weeks, compared with 6-week reports, maternal perception of infant sleep problems was improved for mothers in the SIG, with minimal change for mothers in the CG across the same period. The fact that one half of both groups report their infant’s sleep to be a problem at 3 months suggests ongoing support may be appropriate for parents across the first postnatal year. Future studies should include follow-up at least until 12 months to see if these trends persist.
This study suggests that infant sleep/wake patterns develop of their own accord across the first 3 months postpartum. There were no significant differences in any sleep variables between the infant groups. Total sleep and nocturnal sleep duration increased between 6 and 12 weeks postpartum. Daytime sleep duration and the number of sleep episodes decreased, indicative of sleep consolidation and the maturing of the infant’s sleep/wake system, in line with previous infant sleep research.41,42 These data should help create realistic expectations about early infant sleep development: how often parents might expect their infant to wake during the night, how many sleep periods they may have in 24 hours, the variability between infants, and how long their infant can reasonably be expected to sleep in 1 bout. Increasing understanding of normal infant sleep development may relieve some parental concerns and perceived pressure to intervene with their child’s sleep in this early stage.
Infants seemed to have more sleep in 24 hours when the infant was located outside of the parental bedroom sleep at night. It is not clear why infants in the SIG had less sleep in the parental bedroom; however, it seems that maternal sleep was not adversely affected. Given the greater percentage of infants still in the parental bedroom at night at 3 months, mothers in the SIG might either be more tolerant of their infant’s sleep patterns or less willing to apply commonly encouraged infant self-soothing approaches. Considering current safety recommendations in relation to infant sleep-related deaths are for infants to remain in the parental bedroom for at least the first 6 months,43 this finding warrants further exploration. A transactional sleep framework acknowledges multiple influences on infant sleep development, such as environmental (including health promotion guidelines), cultural (including sleep location), and infant and parental factors (including beliefs, temperament, and choice), meaning approaches that can be tailored to individual families may ultimately have more appeal and increased adherence and use than one-size-fits-all guidelines.44
The current pilot study was necessarily limited by size. Ethnically and demographically similar groups of women were targeted for participation to control those factors. Women already parenting or living in challenging circumstances might find aspects of this methodology more difficult. Keeping track of actigraphs, remembering to press event markers, and completing diaries may be a burden under different conditions. However, participants did not consider the current intervention difficult.
Strengths of this study include the use of a CG and objective measurement of sleep using actigraphy, although we do recognize there are challenges with using and interpreting actigraphy in this young population. Ongoing participant contact contributed to 100% completion of the postnatal phase of the study with all actigraphy data collected being valid for analysis. The education session developed for this study could be adapted to be provided by different means such as face-to-face, via the internet, or as a mobile app with the option of follow-up support calls and adapted for use within different cultural settings.
To date and in general, behavioral–educational sleep interventions have not demonstrated long-term positive effects for maternal sleep or mood. Often studies are limited by participant numbers and/or duration of intervention and follow-up. They also may not target women with a history of or current sleep- or mood-related symptomatology.45 Ideally, future studies would encompass the entire perinatal period, possibly from preconception, although the difficulties of recruiting at this time are acknowledged. Given the ongoing changes to both maternal and infant sleep in pregnancy and the first postnatal year, interventions offered in a program of time-specific modules may be more effective in the short and long term. Ongoing support provided this way may help with intervention adherence and aspects of the cognitive and behavioral therapies (including cognitive behavioral therapy for insomnia46 and mindfulness-based therapy for insomnia47) may strengthen efficacy of approaches targeting sleep and mood. Work should continue to try and understand the mechanisms of changes to sleep and mood at this time including biological (such as hormones, shifts in timing of sleep, and circadian function), social (including cultural and economic influences), and psychological components. Follow-up should also extend well beyond the intervention period to assess durability of effects. Our pilot findings support a positive association between one such module of intervention and greater maternal sleep duration at night, with trends toward increasing sleep quality.
CONCLUSION
Prenatal education and brief postnatal follow-up showed high acceptability and enhanced parental confidence to manage newborn sleep and resulted in an increase in the duration of nocturnal episodes of sleep for mothers in the SIG. These preliminary findings warrant additional research with more diverse and larger samples.
DISCLOSURE STATEMENT
All authors have seen and approved this manuscript for submission. Work for this study was performed at the Sleep/Wake Research Centre, Massey University, Wellington, New Zealand. This study was funded by Health Research Council of New Zealand Grant 09/255. The authors report no conflicts of interest.
ABBREVIATIONS
| CG | control group |
| EPDS | Edinburgh Postnatal Depression Scale |
| GNS | good night’s sleep |
| Noct-TIB | total nocturnal time in bed |
| PIPIS | Parent Information for Parent and Infant Sleep |
| SD | standard deviation |
| SIG | sleep intervention group |
| TIB | time in bed |
| TST | total sleep time |
REFERENCES
1. . Negotiating sleep: a qualitative study of new mothers. J Perinat Neonatal Nurs. 2007;21(2):114–122. https://doi.org/10.1097/01.JPN.0000270628.51122.1d
2. . Sleep patterns and fatigue in new mothers and fathers. Biol Res Nurs. 2004;5(4):311–318. https://doi.org/10.1177/1099800403262142
3. . Child sleep coaches: current state and future directions. Clin Pediatr (Phila). 2017;56(1):5–12. https://doi.org/10.1177/0009922816678977
4. . Infant sleep development from 3 to 6 months postpartum: links with maternal sleep and paternal involvement. Monogr Soc Res Child Dev. 2015;80(1):107–124. https://doi.org/10.1111/mono.12147
5. . A review of pregnancy information on nutrition, physical activity and sleep websites. Women Birth. 2019;93(11):35–40
6. . Randomised controlled trial of behavioural infant sleep intervention to improve infant sleep and maternal mood. BMJ. 2002;324(7345):1062. https://doi.org/10.1136/bmj.324.7345.1062
7. . Improving infant sleep and maternal mental health: a cluster randomised trial. Arch Dis Child. 2007;92(11):952–958. https://doi.org/10.1136/adc.2006.099812
8. . Behavioral interventions for infant sleep problems: a randomized controlled trial. Pediatrics. 2016;137(6):e 20151486. https://doi.org/10.1542/peds.2015-1486
9. . The joy of parenting: infant sleep intervention to improve maternal emotional well-being and infant sleep. Singapore Med J. 2017;58(1):50–54. https://doi.org/10.11622/smedj.2016046
10. . A postpartum sleep and fatigue intervention feasibility pilot study. Behav Sleep Med. 2018;16(2):185–201. https://doi.org/10.1080/15402002.2016.1180523
11. . A behavioral-educational intervention to promote maternal and infant sleep: a pilot randomized, controlled trial. Sleep. 2006;29(12):1609–1615. https://doi.org/10.1093/sleep/29.12.1609
12. . Behavioral sleep interventions in the first six months of life do not improve outcomes for mothers or infants: a systematic review. J Dev Behav Pediatr. 2013;34(7):497–507. https://doi.org/10.1097/DBP.0b013e31829cafa6
13. . Infant sleep disturbance: description of a problem behaviour process. Sleep Med Rev. 1999;3(4):265–280. https://doi.org/10.1053/smrv.1999.0071
14. . Getting rhythm: how do babies do it?. Arch Dis Child Fetal Neonatal Ed. 2015;100(1):F50–F54. https://doi.org/10.1136/archdischild-2014-306104
15. . Comment on: The joy of parenting: infant sleep intervention to improve maternal emotional well-being and infant sleep. Singapore Med J. 2017;58167
16. . Response-based sleep intervention: helping infants sleep without making them cry. Early Hum Dev. 2017;10849–57. https://doi.org/10.1016/j.earlhumdev.2017.03.008
17. . Do women change their health behaviours in pregnancy? Findings from the Southampton Women’s Survey. Paediatr Perinat Epidemiol. 2009;23(5):446–453. https://doi.org/10.1111/j.1365-3016.2009.01036.x
18. . Growing Up in New Zealand: A longitudinal study of New Zealand children and their families. Report 1: Before we are born. Auckland, New Zealand: Growing Up in New Zealand; 2010
19. . Infant sleep problems: origins, assessment, interventions. Infant Ment Health J. 1993;14(1):17–34. https://doi.org/10.1002/1097-0355(199321)14:1<17::AID-IMHJ2280140103>3.0.CO;2-Q
20. . Parenting and infant sleep. Sleep Med Rev. 2010;14(2):89–96. https://doi.org/10.1016/j.smrv.2009.05.003
21. . The prevalence of symptoms of depression and anxiety, and the level of life stress and worry in New Zealand Māori and non-Māori women in late pregnancy. Aust NZ J Psychiatry. 2017;51(2):168–176. https://doi.org/10.1177/0004867415622406
22. . Detection of postnatal depression: development of the 10-item Edinburgh Postnatal Depression Scale. Br J Psychiatry. 1987;150(6):782–786. https://doi.org/10.1192/bjp.150.6.782
23. . Screening for depression during pregnancy with the Edinburgh Depression Scale (EDDS). J Reprod Infant Psychol. 1990;8(2):99–107. https://doi.org/10.1080/02646839008403615
24. . 2007 Sleep in America Poll. Washington, DC: National Sleep Foundation; 2007
25. . Practice parameters for the use of actigraphy in the assessment of sleep and sleep disorders: an update for 2007. Sleep. 2007;30(4):519–529. https://doi.org/10.1093/sleep/30.4.519
26. . The role and validity of actigraphy in sleep medicine: an update. Sleep Med Rev. 2011;15(4):259–267. https://doi.org/10.1016/j.smrv.2010.10.001
27. . Use of actigraphy for assessment in pediatric sleep research. Sleep Med Rev. 2012;16(5):463–475. https://doi.org/10.1016/j.smrv.2011.10.002
28. . The SBSM guide to actigraphy monitoring: clinical and research applications. Behav Sleep Med. 2015;13(suppl 1):S4–S38. https://doi.org/10.1080/15402002.2015.1046356
29. . Immunisation Handbook. Wellington, New Zealand: Ministry of Health: 2002
30. . Sleep measurement in flight crew: comparing actigraphic and subjective estimates to polysomnography. Aviat Space Environ Med. 2005;76(11):1058–1063
31. . Normal sleep in children and adolescents. Child Adolesc Psychiatr Clin. 2009;18(4):799–811. https://doi.org/10.1016/j.chc.2009.04.013
32. . Actigraphic sleep and developmental progress of one-year-old infants. Sleep Biol Rhythms. 2012;10(2):77–83. https://doi.org/10.1111/j.1479-8425.2011.00525.x
33. . Actigraphy correctly predicts sleep behavior in infants who are younger than six months, when compared with polysomnography. Pediatr Res. 2005;58(4):761–765. https://doi.org/10.1203/01.PDR.0000180568.97221.56
34. . Methodological challenges when using actigraphy in research. J Pain Symptom Manage. 2007;36(2):191–199. https://doi.org/10.1016/j.jpainsymman.2007.10.008
35. . Effect of external motion on correspondence between infant actigraphy and maternal diary. Infant Behav Dev. 2009;32(3):340–343. https://doi.org/10.1016/j.infbeh.2009.02.002
36. . The visual scoring of sleep and arousal in infants and children. J Clin Sleep Med. 2007;3(2):201–240. https://doi.org/10.5664/jcsm.26819
37. . Using Multivariate Statistics, 6th ed. Boston, MA: Pearson Education; 2013
38. . SAS® for Mixed Models, 2nd ed. Cary, NC: SAS Institute Inc.; 2006
39. . NZDep2006 Index of Deprivation. Wellington, New Zealand: University of Otago; 2007
40. . Mothers’ postpartum sleep disturbance is associated with the ability to sustain sensitivity toward infants. Sleep Med. 2019;6574–83
41. . The development of circadian rhythms in a human infant. Sleep. 1999;22(3):303–310. https://doi.org/10.1093/sleep/22.3.303
42. . Night waking, sleep-wake organization, and self-soothing in the first year of life. J Dev Behav Pediatr. 2001;22(4):226–233. https://doi.org/10.1097/00004703-200108000-00003
43. SIDS and other sleep-related infant deaths: updated 2016 recommendations for a safe infant sleeping environment. Pediatrics. 2016;138(5):e20162938. https://doi.org/10.1542/peds.2016-2938
44. . Parental regulation of infant sleep: round-the-clock efforts for social synchronization. Infant Ment Health J. 2016;37(3):247–258. https://doi.org/10.1002/imhj.21568
45. . Non-pharmacological interventions for improving postpartum maternal sleep: a systematic review and meta-analysis. Sleep Med Rev. 2018;4187–100. https://doi.org/10.1016/j.smrv.2018.01.005
46. . Cognitive behavioral therapy is effective for insomnia during pregnancy: a randomized controlled trial in an ethnically diverse sample. Sleep. 2018;41(suppl 1):A154–A154. https://doi.org/10.1093/sleep/zsy061.404
47. . Using mindfulness for the treatment of insomnia. Curr Sleep Med Rep. 2017;3(2):57–65. https://doi.org/10.1007/s40675-017-0068-1
