Abstract
Although sleep problems are highly prevalent in infants, the intrinsic and extrinsic factors that influence sleep consolidation and regulation in this age group are not well understood. This study aimed to examine the cross-sectional associations of demographic and parental factors with infant sleep characteristics. Participants were 97 Canadian mother-infant dyads primarily from Edmonton, Alberta. Demographic factors (e.g., infant age), parenting practices (e.g., sleep position, sleep initiation methods), and infants sleep characteristics (e.g., the frequency of nighttime awakenings) were assessed using the Brief Infant Sleep Questionnaire. Maternal sleep characteristics (e.g., nighttime sleep duration) were assessed using Actigraph accelerometers. Infant age (mean = 4.24 ± 2.90) was associated with most infant sleep characteristics. In multiple regression models for infant nighttime sleep duration, after removing influential observations, a negative association for side (vs. prone) sleep position was, respectively, observed. In multiple regression models for the frequency of nighttime awakenings in infants, positive associations for infants falling asleep while feeding (vs. in bed alone) and side (vs. prone) sleep position were consistently observed after removing influential observations. Lower nighttime sleep efficiency (B = − 0.08, 95%CI: − 0.13, − 0.02) and longer nighttime wake after sleep onset (B = 1.03, 95%CI: 0.41, 1.65) in mothers were associated with more frequent nighttime awakenings in infants. After removing influential observations, more frequent nighttime awakenings (B = 0.35; 95%CI: 0.09, 0.61) and longer total sleep duration (B = 0.33, 95%CI: 0.11, 0.55) in mothers were also associated with more frequent nighttime awakenings in infants. Sleep initiation methods with less parental involvement, and more continuous and efficient maternal nighttime sleep, tended to be associated with less interrupted infant sleep.
Keywords: Parenting, Young children, Sleep hygiene, Parental sleep, Postpartum
Introduction
Infancy (0–11 months) is a period of rapid development of the sleep–wake pattern [1, 2]. Since circadian rhythm is immature at birth, the sleep–wake pattern in newborns is driven largely by feeding cues and they usually have multiple sleep episodes distributed throughout a 24-h period [1, 2]. The circadian rhythm normally begins to emerge at around 3 months of age and is well-established by the 6 months when infants tend to have a long stretch of nighttime sleep and a diminishing number of daytime sleep (i.e., naps) [1, 2]. This progressively more consolidated sleep is indicated by shorter and fewer naps, longer nighttime sleep duration, and higher nighttime sleep ratio (i.e., ratio of nighttime sleep duration to total sleep duration) indicate more consolidated sleep [2, 3]. Along with sleep consolidation, infants gradually develop the ability to independently transition between wakefulness and sleep [2]. Crude indicators of a more developed sleep–wake regulation in infants include shorter sleep latency, shorter nighttime wakefulness duration, and fewer nighttime awakenings [1, 2].
While most infants attain consolidated and regulated sleep during the first year of life, sleep problems (e.g., long sleep latency, short nighttime sleep duration, frequent nighttime awakenings) are prevalent in 15–25% of the population in this age group and are particularly pronounced after 6 months of age [3]. These sleep problems in infants may affect physical, cognitive, and socio-emotional development [4] and may track into middle childhood [5]. As such, understanding the factors associated with infant sleep characteristics is needed to facilitate optimal sleep development during infancy.
In a transactional model, Sadeh et al. [6] have emphasized ongoing, bi-directional links between extrinsic (parental) and intrinsic (infant) factors and infant sleep characteristics [6]. According to this model, parental factors have immediate and direct influence on infant sleep [6], which is further supported by empirical evidence on parenting practices in relation to infant sleep characteristics. For example, some studies reported that certain sleep initiation methods (falling asleep with parental intervention vs. independently) [7] and sleep arrangement (co-sleeping [including bed- and room-sharing] vs. solitary-sleeping) [8–10] were associated with increased nighttime awakenings in infants. Placing infants in an appropriate sleep position, particularly before they are able to roll, seems also important for less sleep problems and safety in infants [11]. However, the associations between these parenting practices and other infant sleep characteristics, such as sleep duration, sleep latency, nighttime wakefulness duration, and nighttime sleep ratio, are unclear.
Other parental factors, such as maternal sleep in the postpartum period, seems to be interrelated with infant sleep. It has been reported that women tend to have decreased total sleep duration and sleep efficiency, as well as increased wake after sleep onset (WASO) during early postpartum period [12]. Though previous studies have suggested that these changes in maternal sleep may be predicted by infant sleep, most assessed maternal sleep using subjective measures [12]. Conversely, associations in the opposite direction were observed in two longitudinal studies using a device-based sleep measure (i.e., accelerometer) [13, 14]. The findings of the two studies provided initial evidence that suboptimal maternal sleep may affect infant sleep [13, 14], though this needs to be confirmed, especially in different countries and cultural backgrounds.
The transactional model has also proposed that demographic factors are important for infant sleep development [6]. Empirically, previous studies have reported that some infant sleep characteristics may vary by age (e.g., sleep duration [15]), sex (e.g., sleep duration [16]), and birth order (e.g., sleep problems [17]). However, the associations between these demographic factors and sleep regulation indicators, such as the frequency of nighttime awakenings, are less consistent in infant [2]. Therefore, this study aimed to examine the cross-sectional associations of demographic and parental (parenting practices and maternal sleep characteristics) factors with infant sleep characteristics, in a sample of Canadian mother-infant dyads.
Methods
Study design and participants
The current study is a sub-study of three separate studies (i.e., Prenatal Exercise and Cardiovascular Health (PEACH) study [18], the Metaboreflex (MMHG) study [19], and the physical activity, nutrition and sleep during the postpartum period (SLEEP) study [20]) at the Program for Pregnancy and Postpartum Health between January 2014 and March 2020 (pre-pandemic). More specifically, the PEACH study was a randomized controlled trail, with the purpose of examining the impact of a prenatal exercise intervention on the sympathetic regulation of blood pressure between the second and third trimesters of pregnancy, with a follow up visit at 2 months postpartum [18]. The MMHG study aimed to examine the cross-sectional differences in muscle sympathetic nerve activity and blood pressure between pregnant women in the third trimesters and non-pregnant women, with a follow up visit at 2 months postpartum [19]. The SLEEP study was a cross-sectional study that aimed to examine the relationship between sleep quality and postpartum weight retention [20]. Participants were recruited to the three studies through convenience sampling of women who contacted the Program for Pregnancy and Postpartum Health. Eligibility criteria for participating in the postpartum phase of the PEACH and MMHG studies, as well as in the SLEEP study were: (i) ≥ 18 years, (ii) singleton pregnancy, (iii) between 0 and 52 weeks postpartum, and (iv) capable of reading and writing English. Since data collection procedures were similar in the three studies [18–20], 178 mother-infant dyads took part in postpartum data collection were included in the current study (Fig. 1).
Fig. 1.
Flow diagram for study participants
Measures
Demographic factors (child age, sex, and birth order), parenting practices (sleep arrangement, sleep position, and sleep initiation methods), and infant sleep characteristics (nighttime sleep latency, the frequency of nighttime awakenings, nighttime wakefulness duration [time child spent in wakefulness from 10 in the evening to 6 in the morning], nighttime sleep duration [time child spent in sleep between 7 in the evening and 7 in the morning], nap duration [time child spent in sleep between 7 in the morning and 7 in the evening]) were assessed using the Brief Infant Sleep Questionnaire (BISQ) [21] at a secure website REDCap [22] (in PEACH and MMHG study) or on a hard paper (in SLEEP study). In addition, nighttime sleep duration and nap duration were summed as total sleep duration, and the ratio of nighttime sleep duration from total sleep duration was calculated as nighttime sleep ratio. Maternal sleep characteristics were assessed using Actigraph accelerometers (model wGT3X-BT; Actigraph LLC, Pensacola, FL), which have been validated for sleep assessment in adult population on wrist- and waist-placement [23]. Mothers were asked to wear the accelerometers for seven consecutive days and register ‘time went to sleep’, ‘time woken up from sleep’, ‘nap(s)’, and ‘accelerometer removed’ in activity logs. Participants in the SLEEP study wore the devices on their waist for 24 h/day. In the PEACH and MMHG studies, participants were instructed to wear the devices on their wrist during periods of sleep (night) and on their waist during periods of wakefulness (daytime). Accelerometer data were downloaded to record accelerations over 60-s time intervals (epochs). Non-wear time was identified as periods of ‘accelerometer removed’ that were registered on activity logs. Consistent with a previous study [24], a minimum of three days with ≥ 10 h per day and corresponding nights of data were considered valid for analysis. Sleep and awakening periods were determined using the Actilife software (version 6.13.4; Copyright 2009–2015 Actigraph, LLC) based on the Cole-Kripke algorithm [25] and in consultation with the participant’s activity log. Although the algorithm was initially developed for wrist-worn monitors, a recent study has shown using this algorithm with waist-worn accelerometer can provide valid estimate of sleep characteristics in adults compared with polysomnography [23]. Awakening periods at night of longer than 45 min duration were analyzed as non-sleep activity [25]. During the identified sleep periods, the following measurements were included in the present study: (1) the frequency of nighttime awakening, (2) WASO, (3) nighttime sleep duration, (4) nighttime sleep efficiency, and (5) nap duration. In addition, nighttime sleep duration and nap duration were summed to produce a measure of total sleep duration. The BISQ questionnaire was completed in the same month of the accelerometry, though for the majority of participants (76%) it was completed during the same week of accelerometer wear.
Statistical analysis
Statistical analyses were performed using SPSS version 26.0 (SPSS Inc., Chicago, IL, USA). Descriptive statistics, including means and standard deviations or percentages, were calculated. Linear regression models were performed to examine the associations between demographic and parental factors and infant sleep characteristics. To determine the reference group of multi-categorical variables (i.e., sleep arrangement, sleep position, sleep initiation method) for the linear models, the mean of each category of the variable was estimated from univariate analysis models. Specifically, if infant sleep characteristics were significantly different between one category and others, this category was selected as the reference group for the multi-categorical variable. Each exposure variable was initially included in a simple regression model for each infant sleep characteristic. Exposure variables that met a cut-off of p < 0.10 in the simple regression models were then included in a multiple regression model for the different infant sleep characteristics [26]. If more than one maternal sleep characteristic was significantly associated with an infant sleep characteristic in the simple regression models, separate multiple regression models were conducted to avoid collinearity issues. Observations identified as influential observations based on Cook’s distance values (> 4/n) [27] were removed and models were re-run to check if the results were different. In all simple and multiple regression models, study project (SLEEP, PEACH, MMHG) was included as a covariate to account for unmeasured differences. Statistical significance was set a priori at p < 0.05.
Results
As shown in Fig. 1, of the 178 mother-infant dyads, 31 were excluded from analysis due to not having valid accelerometer data (i.e., maternal sleep characteristics), and 50 were further excluded due to incomplete BISQ data (n = 23; i.e., demographic factors, parenting practices, infant sleep characteristics) or due to the BISQ not being administrated (n = 27). A total of 97 mother-infant dyads had complete data of interest and were included in the analyses. Compared to the analytic sample, the 50 dyads who had valid accelerometer data but not BISQ data were not significantly different in most maternal sleep characteristics, except for having slightly lower nighttime sleep efficiency (88.97 ± 8.28%, p < 0.05). Descriptive data for the analytic sample is presented in Table 1. On average, mothers wore the accelerometer 23.2 ± 1.0 h per day for 6.59 ± 1.19 days, with 92.8% wearing the accelerometer for five days or more.
Table 1.
Descriptive characteristics (n = 97)
| Variables | Mean ± SD or percentage | |
|---|---|---|
| Demographic factors | Child age (month) | 4.24 ± 2.90 |
| Child sex | ||
| Male | 55.7% | |
| Female | 44.3% | |
| Child birth order1 | ||
| Youngest | 37.1% | |
| Oldest | 62.9% | |
| Parenting practice | Sleep arrangement | |
| Solitary-sleeping (infant crib in a separated room) | 44.3% | |
| Room-sharing (infant crib in parents’ room) | 39.2% | |
| Bed-sharing (in parents’ bed) | 16.5% | |
| Sleep position | ||
| Prone (on their belly) | 16.5% | |
| Side (on their side) | 16.5% | |
| Supine (on their back) | 67.0% | |
| Fall asleep | ||
| While feeding | 52.6% | |
| Being rocked | 10.3% | |
| Being held | 14.4% | |
| In bed alone | 14.4% | |
| In bed near parent | 8.2% | |
| Maternal sleep characteristics | The frequency of nighttime awakenings (n) | 1.63 ± 0.97 |
| Nighttime sleep efficiency (%) | 91.48 ± 6.35 | |
| Nighttime WASO (h/d) | 0.73 ± 0.56 | |
| Nighttime sleep duration(h/d) | 7.88 ± 1.06 | |
| Nap duration (h/d) | 0.57 ± 0.71 | |
| Total sleep duration(h/d) | 8.13 ± 1.00 | |
| The number of days with valid accelerometer data (n) | 6.59 ± 1.19 | |
| Infant sleep characteristics | Nighttime sleep latency (h/d) | 0.65 ± 0.55 |
| The frequency of nighttime awakenings (n) | 2.40 ± 1.29 | |
| Nighttime wakefulness duration (h/d)2 | 1.13 ± 1.06 | |
| Nighttime sleep duration (h/d)3 | 9.16 ± 2.02 | |
| Nap duration (h/d)4 | 4.32 ± 1.89 | |
| Total sleep duration (h/d) | 13.48 ± 2.40 | |
| Nighttime sleep ratio5 | 0.68 ± 0.12 | |
WASO wake after sleep onset, h/d hours per day
1The frequency of the category “middle” was zero
2Time child spent in wakefulness from 10 in the evening to 6 in the morning
3Time child spent in sleep between 7 in the evening and 7 in the morning
4Time child spent in sleep between 7 in the morning and 7 in the evening
5Nighttime sleep ratio is the ratio of nighttime sleep duration to total sleep duration
The simple regression models for associations between demographic and parental factors and infant sleep characteristics are presented in Table 2. Age was significantly associated with most infant sleep characteristics, except for the frequency of nighttime awakenings and total sleep duration (p ≥ 0.10). Additionally, several parental factors met the p < 0.10 cut-off in the model for each infant sleep characteristics. Therefore, multiple regression models were conducted and shown in Table 3. Age remained significantly associated with nighttime sleep latency, nighttime wakefulness duration, nap duration (negative), nighttime sleep duration and nighttime sleep ratio (positive) in infants. A few parenting factors and maternal sleep characteristics were associated with the frequency of nighttime awakenings in infants. In the four multiple regression models, there is a consistent pattern of association for sleep initiation method and sleep position after removing influential observations (Model 2, 4, and 5: n = 6; Model 3: n = 7). Specifically, infant who fell asleep while feeding (vs. in bed alone) and who slept in a side position (vs. prone position) had more frequent nighttime awakening. The associations of two maternal sleep characteristics, nighttime sleep efficiency (B = − 0.08; 95%CI: − 0.13, − 0.02) and WASO (B = 1.03; 95%CI: 0.41, 1.65), with the frequency of nighttime awakenings in infants remain significant in the multiple regression models. The association for another two maternal sleep characteristics, the numbers of nighttime awakening (B = 0.35; 95%CI: 0.09, 0.61) and maternal total sleep duration (B = 0.33, 95%CI: 0.11, 0.55), became significant after removing influential observations (Model 2 and 5: n = 6). In the two multiple regression models for infant nighttime sleep duration, the association for sleep position became significant after removing influential observations (Model 7: n = 3; Model 8: n = 5). Specifically, compared to infant who slept in prone position, those who slept in a side position had shorter nighttime sleep duration. Though a difference in nighttime sleep duration was not found between supine and prone position sleepers. For infant total sleep duration, while an association for sleep initiation method (infant fall asleep being held vs. in bed alone) was found, this association ceased to be significant after removing influential observations (Model 11: n = 6).
Table 2.
Simple regression models for the associations between demographic and parental factors and infant sleep characteristics (n=97)
| Exposure variables | Infants sleep characteristics B(95%CI) |
|||||||
|---|---|---|---|---|---|---|---|---|
| Nighttime sleep latency (h/d) | The frequency of nighttime awakenings (n) | Nighttime wakefulness duration (h/d) | Nighttime sleep duration (h/d) | Nap duration (h/d) | Total sleep duration (h/d) | Nighttime sleep ratio (%) | ||
| Demographic factors | Age (months) | − 0.06 (− 0.10, − 0.02) | − 0.04 (− 0.14, 0.06) | − 0.13 (− 0.21, − 0.06) | 0.32 (0.20, 0.44) | − 0.35 (− 0.48, − 0.23) | − 0.03 (− 0.21, 0.15) | 0.03 (0.02, 0.03) |
| Gender (reference group: female) | ||||||||
| Male | − 0.08(− 0.30, 0.14) | 0.25 (− 0.29, 0.78) | 0.27 (− 0.14, 0.69) | − 0.52 (− 1.25, 0.21) | − 0.13 (− 0.89, 0.64) | − 0.65 (− 1.58, 0.28) | − 0.01 (− 0.06, 0.04) | |
| Birth order (reference group: youngest or middle) | ||||||||
| Oldest | 0.08 (− 0.15, 0.30) | 0.17 (− 0.38, 0.72) | 0.01 (− 0.42, 0.44) | − 0.44 (− 1.20, 0.32) | − 0.13 (− 0.92, 0.66) | − 0.57 (− 1.53, 0.39) | − 0.02 (− 0.07, 0.03) | |
| Parenting practices | Sleep arrangement (reference group: bed-sharing) | |||||||
| Solitary-sleeping | − 0.09 (− 0.38, 0.30) | − 0.68 (− 1.43, 0.07)† | 0.31 (− 0.28, 0.89) | 0.37 (− 0.67, 1.42) | − 0.72 (− 1.81, 0.38) | − 0.34 (− 1.68, 1.00) | 0.03 (− 0.03, 0.10) | |
| Room-sharing | − 0.04 (− 0.38, 0.30) | − 0.23 (− 1.03, 0.57) | 0.68 (0.05, 1.30) | − 0.36 (− 1.48, 0.75) | − 0.47 (− 1.64, 0.70) | − 0.84 (− 0.27, 0.59) | 0.003 (− 0.07, 0.07) | |
| Sleep position (reference group: prone) | ||||||||
| Side | − 0.02 (− 0.40, 0.36) | 0.87 (− 0.03, 1.78)† | 0.15 (− 0.57, 0.85) | − 0.76 (− 2.01, 0.49) | 0.002 (− 1.32, 1.32) | − 0.76 (− 2.39, 0.86) | − 0.02 (− 0.10, 0.06) | |
| Supine | 0.15 (− 0.16, 0.45) | 0.34 (− 0.39, 1.06) | 0.51 (− 0.06, 1.08)† | − 1.16 (− 2.17, − 0.16) | 0.63 (− 0.43, 1.69) | − 0.53 (− 1.84, 0.78) | − 0.06 (− 0.13, 0.001)† | |
| Sleep initiation method (reference group: in bed alone) | ||||||||
| While feeding | 0.40 (0.08, 0.71) | 1.05 (0.29, 1.81) | 0.13 (− 0.49, 0.75) | − 1.17 (− 2.23, − 0.10) | − 0.19 (− 1.33, 0.94) | − 1.36 (− 2.71, − 0.01) | − 0.02 (− 0.09, 0.05) | |
| Being rocked | 0.40 (− 0.05, 0.85)† | 0.50 (− 0.58, 1.57) | 0.26 (− 0.63, 1.14) | − 1.66 (− 3.18, − 0.14) | − 0.01 (− 1.61, 1.62) | − 1.65 (− 3.57, 0.27)† | − 0.04 (− 0.14, 0.06) | |
| Being held | 0.36 (− 0.05, 0.76)† | 0.55 (− 0.42, 1.52) | 0.39 (− 0.41, 1.18) | − 1.42 (− 2.79, − 0.05) | − 0.98 (− 2.43, 0.47) | − 2.40 (− 4.12, − 0.67) | 0.01 (− 0.08, 0.09) | |
| In bed near parent | 0.03 (− 0.43, 0.50) | 1.27 (0.15, 2.38) | 0.27 (− 0.64, 1.19) | − 1.13 (− 2.71, 0.44) | − 0.41 (− 2.08, 1.26) | − 1.54 (− 3.52, 0.45) | − 0.01 (− 0.11, 0.10) | |
| Maternal sleep characteristics | Frequency of nighttime awakenings (n) | 0.09 (− 0.04, 0.02) | 0.34 (0.04, 0.64) | 0.05 (− 0.19, 0.29) | − 0.16 (− 0.59, 0.27) | − 0.03 (− 0.42, 0.47) | − 0.13 (− 0.67, 0.41) | 0.002 (− 0.03, 0.03) |
| Nighttime sleep efficiency (%) | − 0.003 (− 0.03, 0.02) | − 0.08 (− 0.13, − 0.03) | − 0.04 (− 0.09, 0.00) | 0.09 (0.01, 0.16) | 0.003 (− 0.08, 0.08) | 0.09 (− 0.01, 0.19)† | 0.003 (− 0.001, 0.01) | |
| Nighttime WASO (h/d) | 0.08 (− 0.19, 0.34) | 1.12 (0.52, 1.72) | 0.41 (− 0.08, 0.91) | − 0.86 (− 1.74, 0.01)† | 0.42 (− 0.50, 1.34) | − 0.44 (− 1.57, 0.69) | − 0.05 (− 0.11, 0.01)† | |
| Nighttime sleep duration (h/d) | 0.07 (− 0.05, 0.18) | 0.17 (− 0.10, 0.44) | − 0.02 (− 0.23, 0.19) | − 0.17 (− 0.54, 0.21) | 0.45 (0.07, 0.83) | 0.28 (− 0.19, 0.76) | − 0.02 (− 0.05, 0.002)† | |
| Nap duration (h/d) | − 0.09 (− 0.25, 0.06) | 0.26 (− 0.12, 0.63) | − 0.01 (− 0.31, 0.28) | − 0.09 (− 0.62, 0.43) | − 0.20 (− 0.74, 0.34) | − 0.29 (− 0.96, 0.37) | 0.01 (− 0.03, 0.04) | |
| Total sleep duration (h/d) | 0.06 (− 0.05, 0.18) | 0.26 (− 0.02, 0.53)† | 0.10 (− 0.12, 0.31) | − 0.33 (− 0.71, 0.06)† | 0.38 (− 0.12, 0.77)† | − 0.05 (− 0.44, 0.54) | − 0.02 (− 0.05, 0.002)† | |
B unstandardized beta coefficient, CI confidence interval, WASO wake after sleep onset; h/d, hours per day
Bold font denotes p < 0.05
†p < 0.10
Each simple regression model was adjusted for study project (SLEEP, PEACH, MMHG)
Table 3.
Multiple regression models for the association between demographic and parental factors and infants sleep characteristics (n = 97)
| Infant sleep characteristics | ||||||||
|---|---|---|---|---|---|---|---|---|
| Exposure variables | Nighttime sleep latency (h/d) | The frequency of nighttime awakenings (n) | Nighttime wakefulness duration (h/d) | – | ||||
| Model 1 | Model 2 | Model 3 | Model 4 | Model 5 | Model 6 | – | ||
| Demographic factors | Age(month) | − 0.05 (− 0.09, − 0.003) | – | – | – | – | -0.12(-0.04,-0.20) | – |
| Parenting practice | Sleep arrangement (reference group: bed-sharing) | |||||||
| Solitary-sleeping | – | − 0.34 (− 1.14,0.45) | − 0.45 (− 1.23,0.34) | − 0.45 (− 1.22,0.32) | − 0.27 (− 1.07,0.54) | 0.36 (− 0.22,0.93) | – | |
| Room-sharing | – | − 0.23 (− 1.05,0.59) | − 0.32 (− 1.12,0.49)a | − 0.30 (− 1.09,0.49)a | − 0.09 (− 0.92,0.73) | 0.41 (− 0.21,1.04) | – | |
| Sleep position (reference group: prone) | ||||||||
| Side | – | 0.76 (− 0.13,1.66)a | 0.64 (− 0.24,1.52)a | 0.63 (− 0.23,1.50)a | 0.77 (− 0.13,1.67)a | 0.13 (− 0.56,0.81) | – | |
| Supine | – | 0.31 (− 0.43, 1.05) | 0.25 (− 0.48,0.97) | 0.23 (− 0.48,0.94) | 0.28 (− 0.46,1.03) | 0.21 (− 0.36,0.77) | – | |
| Sleep initiation method (reference group: in bed alone) | ||||||||
| While feeding | 0.26 (− 0.08,0.60) | 0.71 (− 0.12,1.54)a | 0.73 (− 0.08,1.53)a | 0.61 (− 0.19,1.41)a | 0.77 (− 0.06,1.60)a | – | – | |
| Being rocked | 0.20 (− 0.28,0.67) | 0.21 (− 0.91,1.33) | 0.27 (− 0.82,1.36) | 0.21 (− 0.86,1.28) | 0.26 (− 0.87,1.38) | – | – | |
| Being held | 0.22 (− 0.19,0.64) | 0.22 (− 0.81,1.25) | 0.15 (− 0.86,1.15) | 0.07 (− 0.93,1.06) | 0.41 (− 0.61,1.42) | – | – | |
| In bed near parent | − 0.06 (− 0.53,0.40) | 0.94 (− 0.28,2.15)a | 0.80 (− 0.40,1.99)a | 0.69 (− 0.49,1.87)a | 1.03 (− 0.19,2.24) | – | – | |
| Maternal sleep characteristics | The frequency of nighttime awakenings (n) | – | 0.29 (− 0.02,0.60)a | – | – | – | – | – |
| Nighttime sleep efficiency (%) | – | – | − 0.08 (− 0.13,-0.02) | – | – | − 0.03(− 0.07,0.01) | – | |
| Nighttime WASO (h/d) | – | – | – | 1.03 (0.41,1.65) | – | – | – | |
| Nighttime sleep duration (h/d) | – | – | – | – | – | – | – | |
| Total sleep duration (h/d) | – | – | – | – | 0.21 (− 0.07,0.48)a | – | – | |
| Infant sleep characteristics | ||||||||
|---|---|---|---|---|---|---|---|---|
| Exposure variables | Nighttime sleep duration (h/d) | Nap duration (h/d) | Total sleep duration | Nighttime sleep ratio | ||||
| Model 7 | Model 8 | Model 9 | Model 10 | Model 11 | Model 12 | Model 13 | ||
| Demographic factors | Age(month) | 0.30 (0.15, 0.44) | 0.33 (0.20, 0.46) | − 0.33 (− 0.46, − 0.20) | − 0.34 (− 0.47, − 0.21) | – | 0.03 (0.02, 0.03) | 0.03 (0.02, 0.03) |
| Parenting practice | Sleep arrangement (reference group: bed-sharing) | |||||||
| Solitary-sleeping | – | – | – | – | – | – | – | |
| Room-sharing | – | – | – | – | – | – | – | |
| Sleep position (reference group: prone) | ||||||||
| Side | − 0.85 (− 2.02, 0.32)a | − 0.96 (− 2.13,0.21)a | – | – | – | − 0.03 (− 0.10, 0.04) | − 0.03 (− 0.10, 0.03) | |
| Supine | − 0.62 (− 1.56, 0.32) | − 0.68 (− 1.62,0.27) | – | – | – | − 0.02 (− 0.08, 0.03) | − 0.03 (− 0.08, 0.03) | |
| Sleep initiation method (reference group: in bed alone) | ||||||||
| While feeding | − 0.09 (− 1.16, 0.98) | − 0.11 (− 1.19, 0.97) | – | – | − 0.77 (− 1.91, 0.37) | – | – | |
| Being rocked | − 0.18 (− 1.68, 1.33) | − 0.18 (− 1.70, 1.34) | – | – | − 1.06 (− 2.84, 0.71) | – | – | |
| Being held | − 0.29 (− 1.61, 1.03) | − 0.47 (− 1.78, 0.85) | – | – | − 1.64 (− 3.25, − 0.42)b | – | – | |
| In bed near parent | − 0.33 (− 1.80, 1.13) | − 0.47 (− 1.93, 1.00) | – | – | – | – | – | |
| Maternal sleep characteristics | The frequency of nighttime awakenings (n) | – | – | – | – | – | – | – |
| Nighttime sleep efficiency (%) | 0.05 (− 0.02, 0.12) | – | – | – | 0.07 (− 0.02, 0.17) | – | – | |
| Nighttime WASO (h/d) | – | – | – | – | – | − 0.02 (− 0.07, 0.03) | – | |
| Nighttime sleep duration (h/d) | – | – | 0.23 (− 0.12,0.57) | – | – | – | − 0.004 (− 0.02, 0.02) | |
| Total sleep duration (h/d) | − 0.10 (− 0.46, 0.26) | – | 0.13 (− 0.23, 0.49) | – | – | – | ||
B unstandardized beta coefficient, CI confidence interval, WASO wake after sleep onset; h/d, hours per day
Bold font denotes p < 0.05 (n = 97)
aBecame significant (p < 0.05) when removing influential observations according to Cook’s d values > 4/n
bBecame insignificant (p ≥ 0.05) when removing influential observations according to Cook’s d values > 4/n
Each multiple regression model was adjusted for study projects (SLEEP, PEACH, MMHG)
Discussion
This study examined the associations between demographic and parental factors and infant sleep characteristics. Age was associated with most infant sleep characteristics, while sex and birth order seemed not relevant. Among parenting practices, associations for sleep initiation method (the numbers of nighttime awakening) and sleep position (nighttime sleep duration) were observed. A few sleep characteristics (i.e., the frequency of awakenings, sleep efficiency, and WASO at night; total sleep duration) in mothers were associated with the frequency of nighttime awakening in infants.
Our findings suggest that incorporating age-appropriate methods into bedtime routines to support infants falling asleep independently may facilitate more continuous nighttime sleep in infants. Similarly, previous research assessing infant sleep based on parental report [28] and accelerometry [7] also found unfavorable associations between parental involvement at bedtime and the frequencies of nighttime awakening in infants and toddlers. Together, these findings for sleep initiation methods may be explained by the fact that when less parental dependence is needed for falling asleep, infants have the opportunity to develop internal capacity of self-soothing [2]. This capacity can help infants to stay asleep independently and resume sleep rapidly after physiological arousal occurs [2], which likely leads to more continuous sleep. Literature also suggests that some sleep training strategies (e.g., extinction, positive routine) grounded on principles of learning (e.g., shaping, reinforcement) can help young children develop self-soothing skills. Therefore, these sleep training strategies are thought to be promising for reducing early childhood sleep problems [29, 30]. Ultimately, parents should consult with a pediatrician or other health professional to determine the most appropriate sleep initiation methods based on their child’s age and health status.
The associations observed between sleep positions, the frequency of nighttime awakening, and nighttime sleep duration in our study should not be interpreted as a recommendation to place infants in a prone sleeping position, given evidence on its link with sudden infant death syndrome [31]. Moreover, in contrast to our findings, an American longitudinal study found that infants consistently sleeping in the prone position had more parent-perceived troubling sleeping at 6 months compared to those consistently sleeping in supine and side position [11]. However, the general description of troubling sleeping may not be comparable to the specific sleep characteristics measured in our study, and parental perception of trouble sleeping in young children tends to be inaccurate when compared with screening instruments [32]. Therefore, more evidence on association between sleep position and infant sleep characteristics is needed before a definitive conclusion can be drawn.
Previous studies comparing infant sleep characteristics between co-sleeping and solitary-sleeping arrangements have reported inconsistent findings. For example, a few studies found that parents reported more nighttime awakening in co-sleeping infants [8–10]. However, two longitudinal studies using objective measures (i.e., accelerometry) of sleep in infants reported no differences in frequencies and duration of nighttime awakening between co-sleeping and solitary-sleeping arrangements [33, 34]. Consistent with these two studies, the results from our study further indicate that infant sleep characteristics in solitary-sleeping arrangement may not be different from either type of co-sleeping (i.e., bed- or room-sharing) arrangements. Similar to our findings, a recent longitudinal study observed no differences in accelerometry-based nighttime sleep duration, efficiency, and awakening numbers between persistent room-sharing and persistent solitary-sleeping infants from 3 to 18 months [35]. Although the differences in infant sleep characteristics between room-sharing and bed-sharing arrangements have not been previously examined [35], our findings suggests that such differences are likely due to confounding effects of maternal sleep characteristics. However, given absence of evidence on this topic, this finding needs to be confirmed in future research.
While previous studies suggest infant sleep may be a predictor of self-reported sleep in mothers [12], our findings support the transactional model suggesting that maternal sleep may also be a predictor of infant sleep. Specifically, the associations observed for maternal sleep characteristics may indicate that less continuous and efficient nighttime sleep in mothers may contribute to more frequent nighttime awakenings in infants, independent of parenting practices. Although maternal total sleep duration seemed to favor nighttime awakenings in infants, a longer total sleep duration was likely a compensation of low-quality nighttime sleep in mothers. The associations observed for maternal nighttime sleep characteristics in our study are consistent with the aforementioned two Israel studies [13, 14]. Specifically, a positive association between maternal (at 3 months) and infant (at 6 months) nighttime sleep efficiency was observed in one sample of 57 mother-infant dyads [13]. Trajectories of the associations between maternal and infant nighttime wakefulness from 3 to 18 months were also observed in another sample of 191 mother-infant dyads [14]. One explanation for these associations between maternal and infant nighttime sleep characteristics may be that mothers who have more fragmented nighttime sleep tend to notice infants’ brief awakening [13]. In turn, these mothers may be more frequently involved in settling their children back to sleep at night, which may discourage the infants to develop self-soothing skills and reinforce infants’ nighttime awakening. Alternatively, more disturbed nighttime sleep in mothers may result in poor well-being and less positive parenting [36], which have been found to be associated with infant sleep problems in previous research [37]. These potential mechanisms may need to be further confirmed, and longitudinal studies are warranted to confirm whether bi-directional associations exist between maternal and infant sleep.
In our study, the associations observed between age and nighttime sleep duration, nap duration, and nighttime sleep ratio are consistent with previous studies in infants [8, 16, 38], indicating that infant sleep becomes more consolidated over the first year [2]. Our findings for sleep regulation indicators also align with a few studies, which observed a decrease in nighttime wakefulness duration and sleep latency during the first year [16, 38]. Although evidence on age and frequency of nighttime awakenings has been inconsistent in infants [2], our null finding agrees with a Finnish longitudinal study assessing sleep in children using BISQ [16]. Specifically, this Finnish study reported that the frequencies of nighttime awakenings (2.1–2.5 per night) did not change with age in the first year [16]. As for sex and birth order, although associations have been previously reported in regard to infant sleep characteristics [13, 16, 17], our findings align with others [8, 38], suggesting that these demographic factors may not be relevant.
Strengths of the present study include the examination of multiple infant sleep characteristics and the device-based measure of maternal sleep characteristics. However, this study has several limitations. First, the cross-sectional design precludes causal inference. Second, recall and social desirability biases may have impacted the parental-report measures of infant sleep characteristics. Though, previous research has shown the measures are valid and reliable [21]. Third, the placement of accelerometer during the night were instructed differently between the SLEEP study (waist) and the PEACH study and the MMHG study (wrist), which may lead to measurement differences in maternal sleep characteristics between studies. Though, such differences have been accounted for in the analytic models, as study project was included in the models as a covariate. There may also be unmeasured covariates (e.g., infant temperament, sleep environment, family socioeconomic status, feeding, maternal mental health [6]) leading to residual confounding. Furthermore, data analyzed in the current study were collected from three studies spanning six years, so secular changes in infant and maternal sleep characteristics may exist and impact our findings. However, we have minimized this impact by including study project as a covariate in the analytic models. Finally, the convenience sample utilized for this study may limit the generalizability of our findings to Canadian mother-infant dyads.
Conclusion
Age was the only demographic factor associated with infant sleep characteristics, supporting the various changes that occur with sleep across the first year of life. Supporting infants in falling asleep independently, as well as facilitating continuous and efficient nighttime sleep in mothers, may be important for promoting less frequent nighttime awakening in infants. Further longitudinal research is needed examining the bi-directional associations between maternal and infant sleep. Additionally, future research should also consider paternal or the partner’s sleep, where relevant.
Funding
BAM was funded by WCHRI Summer Studentship. MHD is the Christenson Professor in Active Healthy Living, and a HSFC Joint National and Alberta Improving Hearth Health for Women.New Investigator. VC is supported by a CIHR New Investigator Salary Award. No financial disclosures were reported by the authors.
Declarations
Conflicts of interest
None.
Ethical approval
Ethics Approval for these research studies was obtained from the Research Ethics Board (REB) at the University of Alberta (Pro00043406; Pro00061045; Pro00058560) and conformed to the latest revision of the Declaration of Helsinki, apart from registration in a publicly accessible database. Informed consent was obtained from all participants.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Galland BC, et al. Normal sleep patterns in infants and children: a systematic review of observational studies. Sleep Med Rev. 2012;16(3):213–222. doi: 10.1016/j.smrv.2011.06.001. [DOI] [PubMed] [Google Scholar]
- 2.Camerota M, Propper CB, Teti DM. Intrinsic and extrinsic factors predicting infant sleep: moving beyond main effects. Dev Rev. 2019;53:100871. doi: 10.1016/j.dr.2019.100871. [DOI] [Google Scholar]
- 3.Field T. Infant sleep problems and interventions: a review. Infant Behav Dev. 2017;47:40–53. doi: 10.1016/j.infbeh.2017.02.002. [DOI] [PubMed] [Google Scholar]
- 4.Tham EK, Schneider N, Broekman BF. Infant sleep and its relation with cognition and growth: a narrative review. Nat Sci Sleep. 2017;9:135. doi: 10.2147/NSS.S125992. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Williamson AA, et al. Longitudinal sleep problem trajectories are associated with multiple impairments in child well-being. J Child Psychol Psychiatry. 2020;61(10):1092–1103. doi: 10.1111/jcpp.13303. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Sadeh A, Tikotzky L, Scher A. Parenting and infant sleep. Sleep Med Rev. 2010;14(2):89–96. doi: 10.1016/j.smrv.2009.05.003. [DOI] [PubMed] [Google Scholar]
- 7.Tikotzky L, Sadeh A. Maternal sleep-related cognitions and infant sleep: a longitudinal study from pregnancy through the 1st year. Child Dev. 2009;80(3):860–874. doi: 10.1111/j.1467-8624.2009.01302.x. [DOI] [PubMed] [Google Scholar]
- 8.Bruni O, et al. Longitudinal study of sleep behavior in normal infants during the first year of life. J Clin Sleep Med. 2014;10(10):1119–1127. doi: 10.5664/jcsm.4114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Huang X-N, et al. Feeding methods, sleep arrangement, and infant sleep patterns: a Chinese population-based study. World Journal of Pediatrics. 2016;12(1):66–75. doi: 10.1007/s12519-015-0012-8. [DOI] [PubMed] [Google Scholar]
- 10.Mersky JP, et al. Prevalence and correlates of maternal and infant sleep problems in a low-income US sample. Matern Child Health J. 2020;24(2):196–203. doi: 10.1007/s10995-019-02852-y. [DOI] [PubMed] [Google Scholar]
- 11.Hunt CE, et al. Infant sleep position and associated health outcomes. Arch Pediatr Adolesc Med. 2003;157(5):469–474. doi: 10.1001/archpedi.157.5.469. [DOI] [PubMed] [Google Scholar]
- 12.Hunter LP, Rychnovsky JD, Yount SM. A selective review of maternal sleep characteristics in the postpartum period. J Obstet Gynecol Neonatal Nurs. 2009;38(1):60–68. doi: 10.1111/j.1552-6909.2008.00309.x. [DOI] [PubMed] [Google Scholar]
- 13.Tikotzky L, et al. VII. 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. doi: 10.1111/mono.12147. [DOI] [PubMed] [Google Scholar]
- 14.Tikotzky L, et al. A longitudinal study of the links between maternal and infant nocturnal wakefulness. Sleep Health. 2022;8(1):31–38. doi: 10.1016/j.sleh.2021.09.003. [DOI] [PubMed] [Google Scholar]
- 15.Zhang Z, et al. Demographic correlates of movement behaviors in infants: a longitudinal study. J Phys Act Health. 2022;19(3):177–185. doi: 10.1123/jpah.2021-0570. [DOI] [PubMed] [Google Scholar]
- 16.Paavonen EJ, et al. Normal sleep development in infants: findings from two large birth cohorts. Sleep Med. 2020;69:145–154. doi: 10.1016/j.sleep.2020.01.009. [DOI] [PubMed] [Google Scholar]
- 17.Sadeh A, Mindell J, Rivera L. “My child has a sleep problem”: a cross-cultural comparison of parental definitions. Sleep Med. 2011;12(5):478–482. doi: 10.1016/j.sleep.2010.10.008. [DOI] [PubMed] [Google Scholar]
- 18.Skow RJ, et al. Prenatal exercise and cardiovascular health (PEACH) study: impact on muscle sympathetic nerve (Re)activity. Med Sci Sports Exerc. 2020;53(6):1101–1113. doi: 10.1249/MSS.0000000000002583. [DOI] [PubMed] [Google Scholar]
- 19.Skow RJ, et al. The sympathetic muscle metaboreflex is not different in the third trimester in normotensive pregnant women. J Appl Physiol. 2020;130(3):640–650. doi: 10.1152/japplphysiol.00728.2020. [DOI] [PubMed] [Google Scholar]
- 20.Matenchuk BA, Davenport MH. The influence of sleep quality on weight retention in the postpartum period. Appl Physiol Nutr Metab. 2020;46(1):77–85. doi: 10.1139/apnm-2019-0896. [DOI] [PubMed] [Google Scholar]
- 21.Sadeh A. A brief screening questionnaire for infant sleep problems: validation and findings for an Internet sample. Pediatrics. 2004;113(6):e570–e577. doi: 10.1542/peds.113.6.e570. [DOI] [PubMed] [Google Scholar]
- 22.Harris PA, et al. Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377–381. doi: 10.1016/j.jbi.2008.08.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Full KM, et al. Validation of a physical activity accelerometer device worn on the hip and wrist against polysomnography. Sleep Health. 2018;4(2):209–216. doi: 10.1016/j.sleh.2017.12.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.van der Berg JD, et al. Identifying waking time in 24-h accelerometry data in adults using an automated algorithm. J Sports Sci. 2016;34(19):1867–1873. doi: 10.1080/02640414.2016.1140908. [DOI] [PubMed] [Google Scholar]
- 25.Cole RJ, et al. Automatic sleep/wake identification from wrist activity. Sleep. 1992;15(5):461–469. doi: 10.1093/sleep/15.5.461. [DOI] [PubMed] [Google Scholar]
- 26.Zhang Z, et al. Longitudinal correlates of sleep duration in young children. Sleep Med. 2020;78:128–134. doi: 10.1016/j.sleep.2020.12.023. [DOI] [PubMed] [Google Scholar]
- 27.Cook RD. Influential observations in linear regression. J Am Stat Assoc. 1979;74(365):169–174. doi: 10.1080/01621459.1979.10481634. [DOI] [Google Scholar]
- 28.Mindell JA, et al. Parental behaviors and sleep outcomes in infants and toddlers: a cross-cultural comparison. Sleep Med. 2010;11(4):393–399. doi: 10.1016/j.sleep.2009.11.011. [DOI] [PubMed] [Google Scholar]
- 29.Mindell JA. Empirically supported treatments in pediatric psychology: bedtime refusal and night wakings in young children. J Pediatr Psychol. 1999;24(6):465–481. doi: 10.1093/jpepsy/24.6.465. [DOI] [PubMed] [Google Scholar]
- 30.Mindell JA, et al. Behavioral treatment of bedtime problems and night wakings in infants and young children. Sleep. 2006;29(10):1263–1276. [PubMed] [Google Scholar]
- 31.Dwyer T, et al. Prospective cohort study of prone sleeping position and sudden infant death syndrome. The Lancet. 1991;337(8752):1244–1247. doi: 10.1016/0140-6736(91)92917-Q. [DOI] [PubMed] [Google Scholar]
- 32.Zhang Z, et al. Correlates of nocturnal sleep duration, nocturnal sleep variability, and nocturnal sleep problems in toddlers: results from the GET UP! Study. Sleep Med. 2019;53:124–132. doi: 10.1016/j.sleep.2018.08.035. [DOI] [PubMed] [Google Scholar]
- 33.Teti DM, et al. Sleep arrangements, parent–infant sleep during the first year, and family functioning. Dev Psychol. 2016;52(8):1169. doi: 10.1037/dev0000148. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Volkovich E, et al. Sleep patterns of co-sleeping and solitary sleeping infants and mothers: a longitudinal study. Sleep Med. 2015;16(11):1305–1312. doi: 10.1016/j.sleep.2015.08.016. [DOI] [PubMed] [Google Scholar]
- 35.Volkovich E, et al. Mother–infant sleep patterns and parental functioning of room-sharing and solitary-sleeping families: a longitudinal study from 3 to 18 months. Sleep. 2018;41(2):zsx207. doi: 10.1093/sleep/zsx207. [DOI] [PubMed] [Google Scholar]
- 36.McQuillan ME, et al. Maternal stress, sleep, and parenting. J Fam Psychol. 2019;33(3):349. doi: 10.1037/fam0000516. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Philbrook LE, Teti DM. Bidirectional associations between bedtime parenting and infant sleep: parenting quality, parenting practices, and their interaction. J Fam Psychol. 2016;30(4):431. doi: 10.1037/fam0000198. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Sadeh A, et al. Sleep and sleep ecology in the first 3 years: a web-based study. J Sleep Res. 2009;18(1):60–73. doi: 10.1111/j.1365-2869.2008.00699.x. [DOI] [PubMed] [Google Scholar]