With the birth of every infant the new mother must adjust to the demands of parenting; one component is a requirement to remain “functional” while experiencing potentially severe sleep disruption. Sleep disruption strongly influences daytime functioning including the capacity to deal with the unexpected, changing situations and distractions, as well as the ability to evaluate risks [1]. There is a long history of investigation into the implications of sleep loss and sleepiness of vulnerable populations including shift-workers and commercial drivers [2], [3], however, in comparison postpartum women are largely overlooked.

Childbirth is an extraordinary, everyday experience; in the year 2011, 301 617 infants were born in Australia [4] and 5 229 813 infants were born in the EU [5], resulting in countless potential occurrences of postpartum sleep loss and subsequent daytime sleepiness. Standard sleep research techniques, including sleep diaries and subjective measures of sleepiness have successfully been used to investigate sleepiness in postpartum mothers [6]–[11]. In this series of investigations Insana, Montgomery-Downs and co-authors reported on the interaction between daytime sleepiness, nocturnal sleep and neurobehavioral performance using a combination of subjective and objective measures. In brief, their correlational findings report that daytime sleepiness (measured by the Epworth Sleepiness Scale [ESS]) was lower during postpartum week 13 than in week 7, and lower in postpartum week 7 than week 2 [8]. Despite reduction in ESS over the initial weeks, postpartum women were sleepier compared with control women [7]. In particular, reaction time to the Psychomotor Vigilance Test (PVT) was slower in primiparous women compared with nulliparous control women throughout the first 12 postpartum weeks [7]. Additionally, postpartum women had a significantly shorter sleep onset latency (measured by Multiple Sleep Latency Test [MSLT]) than control women [9]. Interestingly, sleep disruption rather than total sleep obtained was influential in daytime sleepiness [6]. Additionally, subjective sleepiness (measured by Stanford Sleepiness Scale [SSS]) was most associated with sleep quality [8]. The quantity of sleep obtained by new mothers from postpartum weeks 2 to 16 was relatively consistent (7.2 hours). However, sleep quality improved over the same time period due to a reduction in sleep fragmentation and increase in sleep efficiency [10]. Furthermore, while both subjectively and objectively (Actigraphy) recorded sleep times were associated with daytime sleepiness this relationship differed and changed over time [8], demonstrating a unique importance of subjective as well as objective sleep measures.

Throughout the night maternal sleep is regularly interrupted in response to the needs of their infant. Compared to the last month of pregnancy, during the first postpartum month mothers have less night time sleep and spend a greater time awake following sleep onset [12]. This is to be expected as initially infants lack a regular circadian rhythm, which must develop over time, and in many infants is established by postpartum week 12 [13]. At one month postpartum infants may be expected to sleep for an average of 3 to 4.5 hours in a single episode [14]. Despite the fragmentation to maternal sleep there is research evidence to suggest that the total amount of sleep obtained is similar to the general population [10], [15] as new mothers extend their night time sleep period and/or nap during the day. Understanding and mitigating the impact of this sleep disruption is important for the health of the mother, as a large amount of wake after sleep onset and low sleep efficiency are predictive of postpartum fatigue severity and mood in general [16], [17]. The level of sleep disruption experienced is reportedly surprising to new monthers [18], suggesting they may be ill-prepared for its consequences.

Although important evidence regarding the characteristics and implications of sleep disturbance during the postpartum period is starting to emerge, there is still a lack of information regarding longitudinal change during this period and how this impacts daytime sleepiness. In particular, the current study addresses this gap in the scientific knowledge by following the same healthy women over time in a repeated measures design. Previous studies have tended to compare sleep and sleepiness at different postpartum weeks between different groups of mothers or to focus exclusively on those mothers with postpartum depression and/or postpartum fatigue. In addition, the current work considers the postpartum period up to week 18, which is longer than typical in studies of new mothers. This study aimed to quantify longitudinal changes in sleep duration, night time disturbance and daytime sleepiness of a sample of Australian mothers during postpartum weeks 6, 12, and 18. It was hypothesised that; (1) total sleep time would be consistent across time points; (2) sleep disturbance would decrease across time points; (3) daytime sleepiness would be prevalent but decrease across time points; and (4) daytime sleepiness would be correlated with both sleep quantity and sleep quality.

Discussion
The quantity of sleep obtained by healthy new mothers remains consistent across postpartum weeks 6, 12, and 18. With a TST of approximately 7 h 20 m, Australian new mothers obtained more sleep than the average American worker (6 h 53 m) [24]. This is very similar to an American study following new mothers from postpartum weeks 2 to 16, where average TST was 7.2 hours [10]. This provides a cross-cultural replication of the finding that postpartum women experience disturbed sleep, but not necessarily reduced total sleep time. Inevitably, new mothers will wake in the night to attend to their infant, and the number of times per night appears to remain consistent during the first 18 postpartum weeks. This study is one of the first to follow the same healthy new mothers through the first 18 postpartum weeks. Findings are in line with previous findings obtained from comparing different groups of new mothers, where each time point (weeks 2, 7 and 13) was analysed independently [8]. SDI significantly reduced over time, apparently driven by a reduction in WASO minutes, suggesting improved efficiency by mothers at settling their infant back to sleep and/or development of the infant’s circadian rhythm. Furthermore, subjective sleepiness (KSS) at the earliest two time points correlated to SDI but not TST. These two findings highlight the importance of sleep quality as opposed to sleep quantity, especially during the first 12 weeks. In particular, it appears that during the early postpartum period, when sleep disturbance is greatest, the duration of the initial sleep period and longest continuous sleep period are key factors in influencing daytime sleepiness. Later in the postpartum period, when WASO is reduced, TST becomes the sleep measure associated with daytime sleepiness.

Despite the substantial and stable TST, EDS was reported by the majority of participants, with over 50% of participants still reporting EDS at week 18. To contextualise this finding, consider a clinician’s response to an individual presenting with an 18 week history of EDS. Within Australia, the ESS is a recommended tool for assisting clinical decision regarding patient fitness to drive [25]. Using this tool the clinician would likely offer advice regarding implications for driving and daytime impairment. Interestingly, the current study found no correlation between ESS and nocturnal sleep. This is in contrast to a previous regression analysis including ESS during postpartum week 11, where ESS was found to be independently associated with TST[6]. However, participants in the current study reported greater ESS following similar sleep time (Week 12: ESS = 13.3, TST = 439.5 min) compared with the previous study (Week 11: ESS = 8.2, TST = 436.1 min). Little is known about the implications for EDS in healthy postpartum women, however, the clinically relevant amount of EDS within the current study population and reported slowed reaction times of postpartum women compared to controls [7] demonstrate that this is an important area requiring future research.

By week 18, only 5 participants were working full-time. This is likely influenced by Australian legislation which provides parental leave support for up to 18 weeks [26]. EU mothers have similar benefits, being entitled to four months maternity leave [27]. In contrast, American mothers receive 12 weeks unpaid leave [28]. Discrepancies between jurisdictions will likely impact the proportion of mothers returning to the workplace whilst experiencing clinically relevant daytime sleepiness. Despite recognition of the potential dangers resulting from sleepy employees [29] there has been little attempt to understand sleepiness in new parents returning to work. This oversight within the literature has further consequences when considered that despite high levels of sleepiness, new mothers “persevere” to meet their essential work and economical demands [30]. Future research should consider practical implications for high risk activities, such as driving, and the potential impact of self-limiting risk exposure behaviours.

Postpartum mothers are a unique group of otherwise healthy individuals from whom much can be learnt about the effects of sleep disturbance. However, investigations into sleep within the postpartum period have traditionally been restricted to the strong association between sleep and postpartum depression (PPD) [31]–[33]. In particular, sleep duration <6 h in a 24 h period appears associated with PPD [32], and wake after sleep onset and low sleep efficiency are predictive of postpartum fatigue severity [16]. The current findings, with healthy participants, emphasise the importance of sleep disturbance over sleep quantity. With reportedly 16.5% of new mothers experiencing PPD [33] future investigation into SDI within this group could provide valuable understanding.

All data in the current study were collected by self-report. This methodological approach is important because perception of sleep quality in new mothers appears to have a central role in the experience of daytime sleepiness. For example, self-reported sleep quality more consistently accounts for subjective daytime sleepiness than actigraphy recorded TST [8]. Furthermore, there is a stronger association between new mothers subjective sleep quality and mood disturbance than between actigraphy measured TST and mood [17]. However, because all data were collected using a field-based protocol, participants were unsupervised during instrument completion. All participants received clear instructions to complete the diary each day after waking; although, there was no objective facilitation of this. Nonetheless, the field-based design provides ecological validity to results. To minimise participant burden the number of diary entries was kept to as few as possible. Consequently, participants did not regularly record KSS throughout the day; therefore circadian changes cannot be evaluated. At week 18 over half of the participants were still experiencing EDS. Future research should consider an extended study protocol evaluating the prolonged experience of EDS. The small sample size, as well as ethnicity and education level homogeneity limits the generalisability of the research findings. Future research may wish to investigate this topic with a broad range of participants in order to investigate potential differences due to social demographic factors. In addition, future research may wish to consider the role of fathers in infant care and the impact on their sleep and sleepiness. In particular, new fathers who are participating in night time infant care and going to work during the day may be at increased risk of workplace and other accidents due to daytime sleepiness. The first data collection point was during postpartum week 6 because it was considered unethical to request new mothers to participate in this relatively time consuming protocol at any earlier time point. Consequently, the current study provides no information on the first 5 postpartum weeks. As the current work considers only subjective measures of sleep and sleepiness, findings cannot be inferred to objective measures. Future research may also wish to consider the longitudinal relationship between subjective and objective measures of sleep in postpartum mothers.

Using a longitudinal, field-based protocol this study is one of the first to follow new mothers during the initial 18 postpartum weeks. Daytime sleepiness is common and is influenced by night time sleep disturbance (WASO and SDI) rather than total quantity of sleep achieved. EDS can reach clinically significant levels posing implications for safety-critical situations, such as driving. The dynamic nature of sleep during this postpartum period should be taken into account when designing interventions to address sleepiness. For example, a structured remote-learning information program would be useful during the late stages of pregnancy for first time mothers. Such a program could include details about how sleep may be expected to change during the postpartum period and advise on putting coping strategies in place. This would be particularly beneficial as many new mothers are surprised by the level of sleep disruption experienced during this period [18]. Policy makers should consider EDS when determining the minimum parental leave entitlement, and aim to ensure that new parents take adequate time away from the workplace for daytime sleepiness to diminish to a non-critical level.

---

This article is adapted from Ashleigh J. Filtness, Janelle MacKenzie, Kerry Armstrong, Longitudinal Change in Sleep and Daytime Sleepiness in Postpartum Women. Source article. This work is licensed under a Creative Commons Attribution 3.0 License.