A Pilot Randomized, Controlled Trial - Semantic Scholar

A Behavioral-Educational Intervention to Promote Maternal and Infant Sleep: A Pilot Randomized, Controlled Trial Robyn Stremler, RN, PhD1,2; Ellen Hodnett, RN, PhD1; Kathryn Lee, RN, PhD3; Shauna MacMillan, RN, BScN4; Catriona Mill, RN, MHSc2; Lisa Ongcangco, RN, BScN1; Andrew Willan, PhD2 Faculty of Nursing, University of Toronto, Toronto, ON, Canada; 2The Hospital for Sick Children (SickKids), Toronto, ON, Canada; 3Department of Family Health Care Nursing, University of California, San Francisco, CA; 4Queen Elizabeth II Health Sciences Centre, Halifax, NS, Canada

1

Study Objectives: Maternal and infant sleep are significant health concerns for postpartum families. The results of previously published studies have indicated that behavioral-educational strategies promote infant sleep, but these reports relied on parental report and did not include maternal sleep. This pilot study of a maternal-infant sleep intervention evaluated feasibility, acceptability, and effects on sleep and other outcomes in the early postpartum period. Design: Randomized controlled trial with concealed-group allocation. Setting: Hospital postpartum unit with home follow-up. Participants: First-time mothers and their infants randomly assigned to sleep intervention (n = 15) or control group (n = 15). Interventions: The sleep intervention included a 45-minute meeting with a nurse to discuss sleep information and strategies, an 11-page booklet, and weekly phone contact to reinforce information and problem solve. The control group received a 10-minute meeting during which only maternal sleep hygiene and basic information about infant sleep were discussed, a 1-page pamphlet, and calls at weeks 3 and 5 to maintain contact without provision of advice.

Measurement and Results: Questionnaires were completed at baseline and 6 weeks; sleep diaries and mother and infant actigraphy were completed at 6 weeks. The mothers in the sleep intervention group averaged 57 minutes more nighttime sleep, and fewer rated their sleep as a problem, as compared with the mothers in the control group. Infants in the sleep intervention group had fewer nighttime awakenings and had maximum lengths of nighttime sleep that were, on average, 46 minutes longer than those in the control group. Conclusions: A behavioral-educational intervention with first-time mothers in the early postpartum period promotes maternal and infant sleep. Further evaluation of the intervention in a larger, more diverse sample is needed. Keywords: Maternal, infant, sleep, postpartum, behavioral, educational, intervention, randomized controlled trial Citation: Stremler R; Hodnett E; Lee K. A behavioral-educational intervention to promote maternal and infant sleep: a pilot randomized, controlled trial. SLEEP 2006;29(12):1609-1615.

sleep in the early postpartum period.15-18 These studies provided information related to infant sleep and suggested strategies aimed at limiting interactions that could lead to unwanted sleep associations, increasing the infant’s self-soothing ability, and facilitating day-night entrainment. Although these studies found longer, more consolidated sleep periods for those infants who received the experimental intervention, none of the trials offered strategies for improving maternal sleep and none examined the effects of the intervention on maternal sleep or other outcomes. Furthermore, none of these studies used objective measures of sleep to examine infant sleep outcomes. Objective measures are essential to reliable and valid reports of infant sleep patterns.19 Interventions that could improve maternal sleep include sleep hygiene strategies, relaxation techniques,20 development of appropriate expectations related to maternal and infant sleep, and planning to maximize sleep and rest opportunities. Discussion of the challenges of parenting, particularly while sleep deprived, and acknowledgement of the high fatigue level of parenting work may support women in the postpartum period and modulate women’s emotional responses to sleep disruption and fatigue.21 Given that maternal and infant sleep are affected by environmental and social cues and that a mother’s sleep is affected by her infant’s sleep, it seems important that interventions should reflect multiple influences and provide information, strategies, and encouragement to promote sleep for both mother and infant. In preparation for a larger, multicenter randomized, controlled trial, a pilot study was conducted to evaluate a multimodal, behavioral-educational maternal-infant sleep intervention with primiparous women in the early postpartum. The TIPS (Tips for

INTRODUCTION SLEEP IS DIFFICULT TO ACHIEVE, YET IS CRITICALLY IMPORTANT FOR POSTPARTUM FAMILIES. CARING FOR A NEWBORN, WHOSE SLEEP IS DISTRIBUTED ACROSS a 24-hour day and who has frequent nighttime awakenings, means that postpartum women experience sleep deprivation, sleep fragmentation,1,2 and significant fatigue.1,3-5 These alterations to normal adult sleep are particularly pronounced for first-time mothers,6 perhaps due to learning infant care strategies for the first time and adjusting to the maternal role. Infant sleep also presents a challenge; up to 35% of parents report difficulty managing their infant’s night waking or settling,7-11 and there is evidence that infant sleep problems persist into later childhood if not treated.12-14 Clearly, prevention of sleep problems is preferable to treatment of the sequelae of sleep loss and fragmentation. Four randomized controlled trials have examined the effects of behavioral-educational interventions aimed at promoting infant Disclosure Statement This was not an industry supported study. Drs. Stremler, Hodnett, Lee, MacMillan, Mill, Ongcangco, and Willan have indicated no financial conflicts of interest. Submitted for publication May 5, 2006 Accepted for publication August 20, 2006 Address correspondence to: Robyn Stremler, Faculty of Nursing, University of Toronto, Room 288, 155 College Street, Toronto, Ontario, Canada, M5T 1P8; Tel: (416) 978-6925; Fax: (416) 978-8222; E-mail: robyn.stremler@utoronto. ca SLEEP, Vol. 29, No. 12, 2006

1609

An Intervention to Promote Maternal and Infant Sleep—Stremler et al

Infant and Parent Sleep) pilot study evaluated the acceptability and feasibility of various elements central to randomized, controlled trial methods and was designed to obtain sample size and outcome estimates for the planned, multicenter randomized, controlled trial. The TIPS pilot study also investigated the effect of the intervention on infant sleep and maternal sleep, anxiety, depression, and fatigue.

in which mother and baby sleep issues and strategies to improve both were discussed and support and encouragement around the woman’s ability to achieve good sleep for herself and her infant were given. The mother’s partner or other support persons attended the session if they happened to be in the room but were not required to be present. The mother also received an 11-page booklet elaborating on the information discussed, and she was encouraged to refer to the booklet during the first few weeks at home. The aim of the session was to provide the mother with knowledge, skills, and encouragement to address sleep issues in the postpartum period. Topics covered in the session and booklet included maternal sleep hygiene, strategies for increasing maternal sleep, maternal relaxation techniques (e.g., progressive muscle relaxation, deep breathing), acknowledgement of the challenges of parenting and sleep deprivation, information on infant sleep structure and interpreting infant cues, strategies for infant sleep promotion, and strategies for promoting infant night-day entrainment and selfsoothing. During the session, participants were encouraged to express values and opinions and were helped to recognize barriers to implementing the new sleep information. Strategies were tailored to work within or overcome such barriers and to fit with each woman’s beliefs, interests, and lifestyle. For example, women who reported difficulty with or dislike of napping during the day were encouraged to use other ways of increasing total daily sleep time, such as going to bed earlier than usual if the partner was home to look after the baby. The advice given to the sleep intervention group aimed to be culturally sensitive and to respect individual family choices related to infant care, such as room sharing, bed sharing, and infant feeding. The approach in the sleep intervention was to accept these individual family choices, avoid assumptions related to infant care (e.g., that the infant sleep in his or her own room in a crib), and allow for multiple approaches to maternal and infant sleep practices. Mothers were contacted by phone each week (weeks 1-5 after hospital discharge) by the study nurse to reinforce the information given, answer questions, and help the mother to problem solve any sleep difficulties she and her infant were facing. Mothers also had access to the study nurse’s phone number at the research unit; when the study nurse attempted to reach the mother by phone at home, voice mail messages were often left, and the mothers were then able to return the study nurse’s call.

METHODS Setting and Participants Over a 6-week period (September to November 2005), the postpartum unit of a Level III, university-affiliated hospital participated in the study. A maximum of 6 randomizations per week were possible due to the number of actigraphs available, and, because this was a pilot study, recruitment was limited to 6 weeks. Research ethics board approval was obtained prior to commencing the study. Eligible women were those who: had a healthy singleton baby born at gestational age of at least 37 weeks; were first-time mothers, either single or partnered; lived in the greater Toronto area; and were planning to provide fulltime care to their infant for at least the first 6 weeks after discharge home. Families were excluded from participation if the mother or infant had complications requiring prolonged hospital stay, either parent had experienced a previous stillbirth or neonatal death, the mother had a chronic illness that was poorly controlled, the mother used medications that affected sleep, either parent had known drug or alcohol use beyond social use, either parent had a diagnosed sleep disorder, the mother’s partner worked night shifts, the mother was unable to read or understand English, or there was no telephone in the home. Study Procedures Consenting women had baseline data (e.g., birth weight, gestational age, mode of delivery, feeding method) collected from their medical records. Women also filled out a brief questionnaire related to intentions to bed share and room share. Although feeding method and bed sharing have effects on maternal and infant sleep, these were not used as stratification variables because many conversions to bottle feeding and sharing the bed with an infant are reactive responses to challenges with feeding or settling the infant. Similarly, mode of delivery may affect fatigue in the very early postpartum, but these differences do not appear to persist beyond 3 to 4 weeks,22 so this variable was not used for stratification. Women also answered questions related to anxiety, depression, and sleep disturbance. A computer-generated random-number sequence was held by a research associate outside of the immediate research team; women were randomly assigned by paging this centrally controlled randomization service. Randomization occurred as soon as possible after delivery and only if the study nurse was available to deliver the in-hospital component of the assigned intervention. Because this was a small pilot study, there was only 1 study nurse, who had worked as a postpartum nurse and who had received extra training in providing sleep advice to parents. The study nurse provided both the sleep intervention and control interventions.

Control Group Women assigned to the control group received a brief meeting with the study nurse during which only maternal sleep hygiene (e.g., avoidance of caffeine, keeping bedroom cool, dark, and quiet) and basic information about infant sleep (e.g., number of hours a day newborns sleep) was conveyed in a 10-minute meeting. Women received a 1-page pamphlet with this information and a phone call from the study nurse at weeks 3 and 5; these phone calls served to maintain contact, with the aim of increasing compliance with completion of outcome measures at 6 weeks. The study nurse inquired about the woman’s and her baby’s wellbeing but did not offer any advice or support related to maternal or infant sleep. If women asked the study nurse for information related to sleep, the study nurse responded that she was unable to provide such information.

Sleep Intervention Group The sleep intervention included a 45-minute, 1-to-1 meeting SLEEP, Vol. 29, No. 12, 2006

1610


munity who received information from the experimental group. Women were randomly assigned shortly before discharge, with little opportunity to share information with women assigned to the alternate group given the short time period they were in the hospital. To further limit the possibility of contamination, only 1 woman per room was enrolled at a time. It was not possible to achieve blinding of mothers, since the experimental intervention included providing the women with information, strategies, and guidance in implementing strategies to improve sleep. Through the informed consent, the control group knew that they were not receiving as much information related to maternal or infant sleep. Sleep data were collected by actigraphy, which is an objective measure of sleep and unlikely to be affected by disappointment with assignment. Actigraphy data were downloaded, and other self-report data from the women’s sleep diaries and questionnaires were entered by a research assistant who was unaware of group allocation. Autoscoring sleep-analysis software (Action4) also further eliminated potential bias in determining sleep-wake times. Mothers read and responded to the 6-week questionnaires independently and sealed their responses in an envelope for transport to the research unit. In the 6-week questionnaires, women were asked what other sleep-information sources they used.

Both Groups Mothers in both groups were visited in their homes by a research assistant, and actigraphy was performed for approximately 48 hours over 2 consecutive days and nights once the baby reached 6 weeks of age. To accommodate mothers’ schedules and to avoid weekend monitoring, not all babies began monitoring at exactly 6 weeks of age, but all actigraphy was completed before 7 weeks of age were reached. The research assistants were not trained to provide any information on maternal or infant sleep. Each woman wore an Octagonal Basic Motionlogger® (60 grams, Ambulatory Monitoring Inc., Ardsley, NY) actigraph on her nondominant wrist, and each infant wore a MicroMini-Motionlogger® (25 grams, Ambulatory Monitoring Inc.) actigraph around an ankle, over a sock. The actigraph was water resistant but could not be worn while bathing or swimming. Actigraphy provides an objective method of measuring sleepwake parameters in the subjects’ usual environment. The actigraph detects and records continuous motion data by use of a battery-operated wristwatch-size microprocessor that senses motion with a piezoelectric linear accelerometer. These detected movements are translated into digital counts across investigatordesignated recording parameters (1-minute time intervals and 0crossing mode were chosen); the digital counts are then stored in internal memory. Data from the actigraphs are downloaded to a computer using special interface units and then interpreted using autoscoring programs available from the actigraph manufacturer. Action4 software (Ambulatory Monitoring Inc.), which includes scoring algorithms for maternal (Cole-Kripke algorithm) and infant (Sadeh algorithm) data, was used. These algorithms assess the recorded activity for the previous 4 minutes and subsequent 2 minutes before making a determination of sleep or wake for each 1-minute interval. Thus, brief movements in the middle of sleep periods are recorded as sleep, and brief periods of no activity within time intervals of extensive wakeful movement are recorded as wake. Each mother was asked to complete a sleep diary throughout the day and each morning and evening that she and her baby were wearing their actigraphs. The sleep diary asked for information such as bedtimes, wake times, nap times, and feeding times throughout the day, as well as morning and evening information related to fatigue and events relevant to use of the actigraph (e.g., was the actigraph taken off for any reason). Women were also asked to press the event marker on their actigraphs, to indicate when they attempted sleep and when they got up again. Sleep-diary data were used to support the actigraphy data. For example, the sleep diary was used to determine that periods of complete inactivity on the actigraph record corresponded to when the mother recorded that her own or her baby’s actigraph was removed (e.g., for bathing). Following the 2 days and nights of actigraphy and sleep-diary recording, women were again visited by a research assistant who asked them to complete a questionnaire, including ratings of satisfaction with the study, sleep strategies used, and measures of sleep disturbance, anxiety, and depression.

Outcome Measures Maternal and infant sleep outcomes were determined by actigraphy at 6 weeks. There was no baseline measurement of maternal sleep in the immediate postpartum period. Effects of nighttime labor and delivery and the hospital setting prevent measurement of a baseline, and even late third-trimester sleep is disturbed compared with prepregnancy sleep.6 Maternal sleep outcomes included nocturnal (9:00 PM-9:00 AM) sleep time, daytime (9:00 AM-9:00 PM) sleep time, longest nocturnal sleep period without waking, longest daytime sleep period without waking, 24-hour sleep time, wake time after sleep onset at night, and number of nocturnal awakenings. If women forgot to press the event marker on their actigraph to indicate when they tried to sleep and arose from sleeping, those times were manually entered from diary entries; wake time after sleep onset at night was calculated from these self-identified sleep and wake times. Women were also asked if they rated their or their infant’s sleep as “a problem” and if the baby slept in the mother’s bed at any time during the night. Infant sleep outcomes included nocturnal (9:00 PM-9:00 AM) sleep time, daytime (9:00 AM-9:00 PM) sleep time, longest nocturnal sleep period without waking, longest daytime sleep period without waking, 24-hour sleep time, and number of nocturnal awakenings. Actigraphy data for the 2 nights of sleep were averaged for nocturnal sleep outcomes because variables for the 2 nights were significantly correlated. Daytime sleep outcomes were determined from the second day of actigraphy (which provided a full 12 hours of daytime sleep for evaluation), and 24-hour sleep time was determined by adding hours of daytime sleep from the second day to hours of nocturnal sleep from the second night of actigraphy recording. Maternal sleep disturbance was measured using the General Sleep Disturbance Scale at baseline and 6 weeks; the General Sleep Disturbance Scale is a 21-item scale that assesses subjective sleep disturbance over the past 7 days along a number of dimensions, including sleep initiation, sleep maintenance, perceived quality of sleep, and maintenance of wakefulness.23 Women rated the frequency of difficulties in the past week along each dimension of sleep using an 8-point Likert scale from “not

Masking Given the large number of women delivering in the greater Toronto area and the wide geographic area this represents, women in the control group were unlikely to meet someone in their comSLEEP, Vol. 29, No. 12, 2006

1611


and their infants were identified as eligible, and 30 provided consent and were randomly assigned. Many more randomizations could have occurred, but, because of the limited number of actigraphs available, a maximum of 6 families per week were randomly assigned. Of the 34 women who refused participation, 19 were simply not interested in being involved in a research study, 10 felt they were or would be too busy or overwhelmed to participate, 4 were about to be discharged from hospital and did not wish to delay their departure in order to receive the intervention, and 1 woman felt that she would receive adequate advice regarding sleep from family members. Women in the TIPS pilot study were predominantly 30 years of age or older (83%), Caucasian (77%), partnered (100%), and university educated (93%). All women in the study were exclusively breastfeeding at enrollment; breastfeeding initiation rates in Toronto are over 90%.32 Many women had experienced cesarean delivery (47%); this finding may be explained by longer length of stay for women with cesarean delivery and therefore more time for recruitment opportunity. The intervention and control groups

at all” to “everyday.” Internal consistency for the scale is good (Cronbach α reliability coefficient = 0.88). The scale yields a total score ranging from 0 to 147, with higher scores indicating greater frequency of sleep disturbance. A cutoff score of 42 for the total scale distinguishes good and poor sleepers.23 The General Sleep Disturbance Scale was selected over the Pittsburgh Sleep Quality Index, which asks about sleep during the past month and has no established validity or reliability with pregnant or postpartum women. Morning and evening fatigue were measured by the Brief Fatigue Visual Analogue Scale at 6 weeks.24 The 7-item version of the longer 18-item Fatigue Visual Analog Scale consists of 100mm visual analog lines; the line anchors are adjectives that relate to level of perceived fatigue (e.g., not at all tired to extremely tired). This shorter version of the Fatigue Visual Analog Scale has good internal consistency reliability (Cronbach α coefficient = 0.94)25 and has now been used in numerous studies of sleep in childbearing women.1,2,26 Women completed this measure before going to sleep at night and upon awakening in the morning, along with their sleep diary during actigraphy data collection at 6 weeks. To examine depressive symptoms, the Edinburgh Postnatal Depression Scale (EPDS),27 a 10-item self-report instrument, was administered at baseline and 6 weeks. Items inquire about maternal mood in the past 7 days and are rated on a 4-point scale to produce a total score ranging from 0 to 30, with higher scores indicating lower maternal mood. The EPDS does not diagnose postpartum depression; rather, it is the most frequently used instrument to assess for postpartum depressive symptomatology.28 The EPDS has been validated by standardized psychiatric interviews with large samples and has well-documented reliability and validity in more than 11 languages.29 Although EPDS scores greater than 12 suggest the greatest risk for development of postpartum depression, we used the EPDS as a continuous measure of depressive symptomatology. The 20-item self-report State-Trait Anxiety Inventory, stateanxiety subscale30 was administered at baseline and 6 weeks to assess levels of relatively transient, situation-related (state) anxiety. Items are rated on a 4-point Likert-type scale to produce a total score ranging from 20 to 80, with higher scores indicating higher levels of state anxiety. The State-Trait Anxiety Inventorystate subscale has been used widely in pregnancy and postpartum studies, and anxiety has been consistently related to depressive symptomatology in new mothers.28 In a study of an intervention to prevent postpartum depression, Cronbach α coefficients for the state-anxiety subscale at 12 and 16 weeks postpartum were 0.94 and 0.95, respectively.31 Statistical Methods Data were entered into Microsoft Access® 2002 (Microsoft Corporation, Redmond, WA) using double-data entry with builtin logic and range checks and analyzed in SAS Version 9.1 (SAS Institute Inc., Cary, NC). As per the “intention-to-treat” approach, data from all participants in the TIPS pilot study were included in the final analysis. Results are presented as frequencies and means, and the Type I error rate was set at .05 (2 sided) for statistical significance. Dichotomous variables (e.g., exclusive breastfeeding) were analyzed using a contingency table χ2. All continuous variables (e.g., sleep outcomes, depression scores) were analyzed using unpaired 2-sample t tests.

Table 1—Baseline Characteristics of the Women Who Were Randomly Assigned to Groups Sleep Intervention group (n = 15)

Gestational age, wksdays 370-406 12 (80) 12 (80) ≥ 410 3 (20) 3 (20) Mean maternal age, 31.1 ± 3.5 (24-37) 32.6 ± 3.5 (27-39) y, ± SD (range) Highest education level Secondary 2 (13) 0 (0) Postsecondary 13 (87) 15 (100) Married/stable relationship 15 (100) 15 (100) Race Caucasian 12 (80) 11 (73) Asian 2 (13) 3 (20) Hispanic 1 (7) 0 (0) Other 0 (0) 1 (7) Type of delivery Spontaneous vaginal 5 (33) 6 (40) Assisted vaginal 2 (13) 3 (20) Cesarean section 8 (53) 6 (40) Planning to breastfeed 15 (100) 15 (100) Planning to room share 14 (93) 13 (87) Planning to bed share 4 (27) 4 (27) GSDS score Mean ± SD (range) 56.4 ± 16.8 (16-76) 52.0 ± 14.1 (24-77) > 42, poor sleeper 13 (87) 13 (87) STAI score Mean ± SD (range) 35.3 ± 10.3 (24-60) 38.6 ± 10.4 (28-59) 40-59 2 (13) 6 (40) ≥ 60 1 (7) 0 (0) EPDS score Mean ± SD (range) 6.2 ± 4.3 (1-14) 6.6 (4.7, 0-17) > 12 2 (13) 1 (7) Data are presented as number (percentage), unless otherwise indicated. GSDS refers to General Sleep Disturbance Scale (score range, 0-147, a score > 42 indicates poor sleeper); EPDS, Edinburgh Postnatal Depression Scale (score range 0-30, a score > 12 indicates at risk for postpartum depression); STAI State-Trait Anxiety Inventory, state subscale (score range, 20-80, a score 40-59 indicates moderate anxiety; a score ≥ 60, severe anxiety).

RESULTS In 6 weeks of recruitment to the TIPS pilot study, 64 women SLEEP, Vol. 29, No. 12, 2006

Control group (n = 15)

1612


Table 2—Comparison of Maternal Outcomes Between Treatment Groups at 6 Weeks

GSDS score GSDS score > 42 EPDS score EPDS score > 12 STAI STAI score 40-59 Nocturnal sleep, min Longest nocturnal sleep period, min Nocturnal awakenings, no. WASO, min Daytime sleep, min Longest daytime sleep period, min 24-h sleep, min Fatigue–morning, 100mm Fatigue–evening, 100mm Rated own sleep as problem Rated infant sleep as problem Exclusive breastfeeding Bed sharing with infant at some point during night

Sleep Intervention Group n = 15 37.0 (28.7, 45.2) 6 (40%) 4.4 (2.9, 5.9) 0 (0%) 27.1 (23.0, 31.1) 2 (13%) 433 (402, 463) 186 (157, 215) 11.2 (8.5, 13.9) 126 (96, 157) 68 (27, 109) 27 (5, 48) 497 (438, 556) 36.5 (27.1, 45.9) 57.5 (49.1, 66.0) 5 (33%) 6 (40%) 8 (53%) 5 (33%)

Control Group n = 15 45.9 (35.9, 55.9) 8 (53%) 5.9 (4.0, 7.9) 0 (0%) 29.1 (24.5, 33.6) 2 (13%) 376a (332, 420) 158a (127, 189) (8.9, 16.5) 12.7a 116a (88, 144) 78 (39, 118) 42 (10, 75) 462 (404, 520) 48.4 (36.5, 60.4) 62.3 (55.6, 69.0) 11 (73%) 9 (60%) 10 (67%) 7 (47%)

Between treatmentp Value group difference -8.9 (-21.3, 3.5) .15 χ2 = 0.54 .46 -1.5 (-3.9, 0.8) .19 1.00 χ2 = 0 -2.0 (-7.8, 3.8) .49 1.00 χ2 = 0 57 (6, 107) .03 28 (-12, 68) .17 -1.5 (-5.9, 2.9) .49 11 (-29, 50) .59 -10 (-65, 44) .69 -15 (-52, 21) .39 35 (-44, 114) .38 -11.9 (-26.3, 2.4) .10 -4.7 (-15.1, 5.7) .36 χ2 = 4.82 .03 χ2 = 1.20 .27 χ2 = 0.56 .46 .46 χ2 = 0.56

Values are means (95% CI) or number (%), unless otherwise indicated. GSDS refers to General Sleep Disturbance Scale (score range, 0-147, a score >42 indicates poor sleeper); EPDS, Edinburgh Postnatal Depression Scale (score range 0-30, score > 12 indicates at risk for postpartum depression); STAI, State-Trait Anxiety Inventory state subscale (score range, 20-80; a score of 40-59 indicates moderate anxiety; a score ≥ 60, severe anxiety). a n = 14. Table 3—Comparison of Infant Outcomes Between Treatment Groups at 6 Weeks

Nocturnal sleep, min Longest nocturnal sleep period, min Nocturnal awakenings, no. Daytime sleep, min Longest daytime sleep period, min 24-h sleep, min

Sleep Intervention Group n = 15 468 (445, 491) 217 (189, 245) 7.9 (5.5, 10.2) 231 (197, 265) 76 (58, 94) 697 (656, 739)

Control Group n = 15 448 (418, 478) 171 (138, 204) 12.3 (10.1, 14.6) 276 (238, 314) 100 (69, 131) 728 (675, 782)

Between treatmentp Value group difference 20 (-16, 56) .26 46 (5, 88) .03 -4.4 (-7.6, -1.4) .006 -45 (-93, 4) .07 -24 (-58, 11) .17 -31 (-96, 33) .33

Values are means (95% CI) unless otherwise indicated.

baby would fit into these patterns. Ten women (67%) in the sleep intervention group received 4 or 5 phone calls, 3 women (20%) received 3 calls, and 2 women (13%) received 2 calls for support and problem solving prior to outcomes determination at 6 weeks. Thirteen women (87%) in the control group received 2 phone calls, and 2 women received 1 call to maintain rapport. All 30 women participated in the home visits at 6 weeks. In cases in which actigraphy data were lost due to equipment failure (2 baby, 1 mother), sleep-diary data were substituted. Only 3 women (10%) had incomplete sleep diaries. Sleep outcome data were missing for 1 woman who did not record her sleep and for whom actigraphy data also were not available due to technical difficulties. Women in the sleep intervention group used all of the strategies and found many of them helpful. Among all the participants, 20 women (67%) asked other mothers for advice on their infant’s sleep, whereas only 6 (20%) asked a physician for information. Seventeen women (57%) consulted books, and 7 women referred to the Internet (23%) for information on sleep. There were no differences between groups in terms of use of other sources of sleep

were similar on baseline and demographic variables (Table 1). Outcomes At 6 weeks postpartum, women in the sleep intervention group had more nighttime sleep (433 vs 376 minutes, group difference 57 minutes [95% confidence interval {CI}: 6-106 minutes], 2p = .03) and fewer rated their sleep as a problem (χ2 = 4.82, 2p = 0.03). Infants in the sleep intervention group had fewer nighttime awakenings (7.9 vs 12.3, group difference 4.4 [95% CI: 1.4-7.6], 2p = .006) and longer maximum lengths of nighttime sleep (217 vs 171 minutes, group difference 46 minutes [95% CI: 5-88 minutes], 2p = .03). No statistically significant differences were found between groups on the other outcomes (Table 2 and Table 3). Adherence to and Acceptability of the Intervention and Data Collection All women completed the assigned in-hospital intervention prior to discharge. Women in the sleep intervention group actively participated, sharing their sleep habits and considering how their SLEEP, Vol. 29, No. 12, 2006

1613


information. All but 1 woman indicated that they would participate in the study if they had to do it over again; this woman was assigned to the control group.

the older infant who should be sleeping through the entire night. Women in the control group received less attention than women in the experimental group. However, there is no research evidence of links between attention given by health care professionals and improvements in maternal or infant sleep and no theoretical reason to suspect such a relationship. Furthermore, the primary and secondary sleep outcomes were measured objectively using actigraphy. Because the sleep intervention contained numerous strategies aimed at promoting maternal and infant sleep and all women in the experimental group used most of the components of the intervention, we are unable to determine if certain strategies were more useful than others. Similarly, we don’t know if the behavioral-educational session delivered to women while in the hospital was the more “active ingredient” of the intervention, if more benefit was derived from receiving follow-up phone calls and assistance with problem solving, or if the components had an additive effect. Because maternal and infant sleep in the early postpartum period are affected by a number of factors, it seems important to maintain multiple strategies in future sleep intervention studies. Although the findings of the TIPS pilot study are encouraging, they are unlikely to change current practice, given the small homogeneous sample and wide confidence intervals around group differences. It will be essential to conduct a larger trial to detect differences on important outcomes such as maternal fatigue and depression, enroll women from diverse cultural and socioeconomic backgrounds, and examine outcomes at ages when infants are physiologically able to sleep for a longer uninterrupted period without a feeding.

DISCUSSION The sleep intervention in the TIPS pilot study was effective in promoting maternal and infant sleep in the early postpartum period, as evidenced by increased maternal nighttime sleep time and longer infant nighttime sleep periods with fewer infant nighttime awakenings. The clinical significance of the effects on maternal sleep is evidenced by fewer women in the sleep intervention group rating their sleep as a problem. This is the first randomized controlled trial to evaluate interventions to promote maternal sleep, and the first to find evidence of the effectiveness of such strategies. Data from 4 previously conducted randomized controlled trials provide evidence of efficacy of strategies aimed at promotion of infant sleep in the early postpartum.15-18 The findings from the TIPS pilot study give similar evidence of the effectiveness of strategies aimed at understanding infant cues for sleep, promoting infant self-soothing, and entraining infants to day-night differentiation; our findings are strengthened by the use of actigraphy for the evaluation of sleep outcomes. Congruence between polysomnography and actigraphy indicates adequate validity and reliability when sleep is assessed in healthy young adults, including women of childbearing age,33,34 with 88% agreement between the two methodologies. Actigraphy has also been used successfully to study sleep-wake patterns in newborn and older infants19,35,36 and has been shown to have an overall 95% minute-by-minute agreement with direct observation of infant sleep-wake states.37 Previously conducted actigraphy studies have recommended 7 days of recording to determine night-to-night variability, but, with pregnant women and new parents, 48 hours during weekdays is commonly used to limit demands on already challenged families and to decrease systematic error from recording over weekends when sleep patterns change drastically for young adults.26 A limitation of this pilot work is the use of 1 nurse to deliver both the experimental and control interventions. Although the study protocol strictly defined which information and strategies were to be provided to each group, it is possible that extra sleep information was given to the control group. However, if such contamination had occurred between groups, it would only be expected to dilute any difference between groups, not exaggerate it. Other sources of sleep information used were equally distributed between the 2 groups, and the effects of co-intervention are therefore expected to be minimal. Usual care for postpartum women is to receive no additional information regarding maternal and infant sleep, either in the hospital or through contact with public health or primary care providers after discharge. Furthermore, health care professionals receive little content on sleep during their academic education.38,39 In Toronto, public health nurses do not receive any additional training in maternal and infant sleep; assessment is made primarily for excessive sleepiness in the infant, because this may compromise infant feeding, and inability to achieve any sleep in the mother, because this may indicate postpartum depression. The most readily available information for parents related to sleep, either from health care professionals, reading materials, or online information focuses not on promotion of maternal or infant sleep in the early postpartum, but on treating SLEEP, Vol. 29, No. 12, 2006

ACKNOWLEDGMENTS Faculty of Nursing, University of Toronto and The Hospital for Sick Children (SickKids), Toronto, Ontario, Canada. We are indebted to the families who participated in the study. We also thank the staff of the Postpartum Unit of Sunnybrook Health Sciences Centre and Janet Yamada for their assistance in carrying out this work. At the time of the study, Dr. Stremler was supported through a salary award from the Randomized Controlled Trials Mentoring Program at the Canadian Institutes of Health Research. REFERENCE 1. 2. 3. 4. 5. 6. 7. 8. 1614

Lee KA, DeJoseph JF. Sleep disturbances, vitality, and fatigue among a select group of employed childbearing women. Birth 1992;19:208-13. Lee KA, Zaffke ME. Longitudinal changes in fatigue and energy during pregnancy and the postpartum period. J Obstet Gynecol Neonatal Nurs 1999;28:183-91. Fawcett J, York R. Spouses’ physical and psychological symptoms during pregnancy and the postpartum. Nurs Res 1986;35:144-8. Gjerdingen DK, Froberg DG, Chaloner KM, McGovern PM. Changes in women’s physical health during the first postpartum year. Arch Fam Med 1993;2:277-83. Harrison MJ, Hicks SA. Postpartum concerns of mothers and their sources of help. Can J Public Health 1983;74:325-8. Lee KA. Alterations in sleep during pregnancy and postpartum: a review of 30 years of research. Sleep Med Rev 1998;2:231-42. Ferber R. Sleep, sleeplessness, and sleep disruptions in infants and young children. Ann Clin Res 1985;17:227-34. Hiscock H, Wake M. Randomised controlled trial of behavioural infant sleep intervention to improve infant sleep and maternal mood. An Intervention to Promote Maternal and Infant Sleep—Stremler et al

9. 10. 11. 12. 13. 14. 15. 16. 17.

18. 19. 20. 21.

22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34.

BMJ 2002;324:1062-5. Ramchandani P, Wiggs L, Webb V, Stores G. A systematic review of treatments for settling problems and night waking in young children. BMJ 2000;320:209-13. Johnson CM. Infant and toddler sleep: a telephone survey of parents in one community. J Dev Behav Pediatr 1991;12:108-14. Anders TF. Night-waking in infants during the first year of life. Pediatrics 1979;63:860-4. Kataria S, Swanson MS, Trevathan GE. Persistence of sleep disturbances in preschool children. J Pediatr 1987;110:642-6. Zuckerman B, Stevenson J, Bailey V. Sleep problems in early childhood: continuities, predictive factors, and behavioral correlates. Pediatrics 1987;80:664-71. Jenkins S, Owen C, Bax M, Hart H. Continuities of common behaviour problems in preschool children. J Child Psychol Psychiatry 1984;25:75-89. Pinilla T, Birch LL. Help me make it through the night: behavioral entrainment of breast-fed infants’ sleep patterns. Pediatrics 1993;91:436-44. Wolfson A, Lacks P, Futterman A. Effects of parent training on infant sleeping patterns, parents’ stress, and perceived parental competence. J Consult Clin Psychol 1992;60:41-8. St James-Roberts I, Sleep J, Morris S, Owen C, Gillham P. Use of a behavioural programme in the first 3 months to prevent infant crying and sleeping problems. J Paediatr Child Health 2001;37:28997. Symon BG, Marley JE, Martin AJ, Norman ER. Effect of a consultation teaching behaviour modification on sleep performance in infants: a randomised controlled trial. Med J Aust 2005;182:215-8. Sadeh A. Evaluating night wakings in sleep-disturbed infants: a methodological study of parental reports and actigraphy. Sleep 1996;19:757-62. McQueen A, Mander R. Tiredness and fatigue in the postnatal period. J Adv Nurs 2003;42:463-9. Fisher J, Feekery C, Rowe H. Treatment of maternal mood disorder and infant behaviour disturbance in an Australian private mothercraft unit: a follow-up study. Arch Women Ment Health 2004;7:8993. Lee KA, Zaffke ME, McEnany G. Parity and sleep patterns during and after pregnancy. Obstet Gynecol 2000;95:14-8. Lee KA. Self-reported sleep disturbances in employed women. Sleep 1992;15:493-8. Lee KA, Hicks G, Nino-Murcia G. Validity and reliability of a scale to assess fatigue. Psychiatry Res 1991;36:291-8. Waters MA, Lee KA. Differences between primigravidae and multigravidae mothers in sleep disturbances, fatigue, and functional status. J Nurse Midwifery 1996;41(5):364-7. Lee KA, Gay CL. Sleep in late pregnancy predicts length of labor and type of delivery. Am J Obstet Gynecol 2004;191(6):2041-6. Cox JL, Holden JM, Sagovsky R. Detection of postnatal depression. Development of the 10-item Edinburgh Postnatal Depression Scale. Br J Psychiatry 1987;150:782-6. Beck CT. Predictors of postpartum depression: an update. Nurs Res 2001;50(5):275-85. Affonso DD, De AK, Horowitz JA, Mayberry LJ. An international study exploring levels of postpartum depressive symptomatology. J Psychosom Res 2000;49:207-16. Spielberger CD. STAI manual for the state-trait anxiety inventory. Palo Alto: Consulting Psychologists Press; 1970. Dennis C. The effect of peer support on postpartum depression: a pilot randomized controlled trial. Can J Psychiatry 2003;48:61-70. Toronto perinatal and child health survey. Toronto: Toronto Public Health; 2003. Ancoli-Israel S, Cole R, Alessi C, Chambers M, Moorcroft W, Pollak CP. The role of actigraphy in the study of sleep and circadian rhythms. Sleep 2003;26:342-92. Walsh JK, Schweitzer PK, Anch AM, Muehlbach MJ, Jenkins NA,

SLEEP, Vol. 29, No. 12, 2006

35. 36.

37. 38. 39.

1615

Dickins QS. Sleepiness/alertness on a simulated night shift following sleep at home with triazolam. Sleep 1991;14:140-6. Acebo C, Sadeh A, Seifer R, et al. Estimating sleep patterns with activity monitoring in children and adolescents: how many nights are necessary for reliable measures? Sleep 1999;22:95-103. Sadeh A, Lavie P, Scher A, Tirosh E, Epstein R. Actigraphic homemonitoring sleep-disturbed and control infants and young children: a new method for pediatric assessment of sleep-wake patterns. Pediatrics 1991;87:494-9. Sadeh A, Hauri PJ, Kripke DF, Lavie P. The role of actigraphy in the evaluation of sleep disorders. Sleep 1995;18:288-302. Mindell JA, Moline ML, Zendell SM, Brown LW, Fry JM. Pediatricians and sleep disorders: training and practice. Pediatrics 1994;94:194-200. Mindell JA, Owens JA. Sleep problems in pediatric practice: clinical issues for the pediatric nurse practitioner. J Pediatr Health Care 2003;17:324-31.
