One in six United States (U.S.) children between the ages of 3–17 have developmental disabilities (DDs), life-long conditions that have significant limitations in language, self-care, learning, and/or mobility (Developmental Disabilities Assistance and Bill of Rights Act of 2000; Zablotsky et al., 2019). Mothers are typically the primary caregivers for these children (Lee, 2013) and commonly report impaired sleep that has been characterized by short sleep duration, more sleep disruptions, and poor sleep quality compared to mothers of typically developing children (Gallagher et al., 2009b; Hodge et al., 2013; Meltzer, 2008; Micsinszki et al., 2018). Impaired sleep is consistently linked to poor mental and physical health, such as depression (Buysse et al., 2008), cardiovascular disease (Kato et al., 2009), obesity (Nielsen et al., 2011), and type 2 diabetes (Cappuccio et al., 2010). Mothers of children with DDs have reported poor mental and physical health, such as depression (Gallagher & Hannigan, 2014), hypertension (Gallagher & Whiteley, 2012), altered immune functions (Gallagher et al., 2009a), and elevated concentrations of proinflammatory biomarkers (Lovell et al., 2012). One population-based longitudinal study reported that mothers who lived with their adult children with DDs had greater body mass index and impaired cardiovascular and daily functioning in their mid-60s compared to the age-matched comparison group (Seltzer et al., 2011), which underscores the negative impact of lifelong caregiving on these mothers’ health especially later in life. Sleep is a modifiable health behavior critical to one’s health and well-being, and current evidence shows that mothers of children with DDs experience impaired sleep. However, existing research relies heavily on mothers’ self-reported sleep.

Relying on self-report may be problematic as 14.9 to 52.5% of these mothers often experience high depressive symptoms (Gallagher et al., 2009b; Lee et al., 2018a; Wayte et al., 2012) may alter perceptions of their and their children’s sleep problems. Furthermore, although mothers’ impaired sleep was associated with their child’s sleep problems and nighttime caregiving activities for their child with DDs (Bourke-Taylor et al., 2013; Goldman et al., 2014; Lee et al., 2018b; Wayte et al., 2012), available studies were unable to delineate the extent to which child sleep-wake behaviors required mothers’ nighttime caregiving and its influence on mothers’ impaired sleep. Actigraphy is widely used for objectively recording sleep from small, watch-like devices worn on the body (Sadeh, 2011). Videosomnography has been used as a non-invasive sleep measure that allows observation of the child’s sleep and caregivers’ activities in their home (Schwichtenberg et al., 2018). Simultaneous recording of mothers’ sleep using actigraphy and children with DDs’ sleep using videosomnography allowed us to objectively obtain child sleep-wake behaviors and identify which child sleep-wake behaviors require mothers’ nighttime caregiving. We also obtained well-known risk factors for mothers’ impaired sleep, such as depressive symptoms (Gallager & Hannigan, 2014; Lee et al., 2018a) and caregiving stress (Lee, 2013), in this study. This study provided preliminary information on the relative contributions of the child’s sleep problems and other risk factors on mothers’ impaired sleep that can be the foundation for a tailored sleep intervention in this population.

Our focus is on mothers of school-aged children as these mothers have been in the caregiving role for years and often have years of future caregiving. The specific aims of this study were to assess the feasibility (recruitment, adherence, acceptability, and missing data) of using (1) actigraphy to characterize mothers’ sleep-wake variables (total sleep time [TST], wake after sleep onset [WASO], number of night wakings, sleep efficiency, duration, and frequency of naps); (2) videosomnography to obtain child’s sleep-wake variables (sleep onset/offset time, WASO, the number and reasons of night waking) including mothers’ nighttime caregiving activities; and; (3) obtain preliminary data on sleep-related variables (perceived sleep quality, depressive symptoms, caregiver stress, and child sleep problems) in mothers of school-aged children with DDs.

Methods

Study Design, Settings, and Sample

This was an observational pilot study with prospective data collection for 7 consecutive days/nights. A convenience sample of mothers and their school-aged children with DDs were recruited from a non-profit community organization (which provides therapeutic play sessions to school-aged children with DDs) that serves the metropolitan area of a southeastern city. Successful recruitment strategies used in our previous study (Lee & Spratling, 2019) were included: (1) play session leaders and staff referring potential participants to the principal investigator (PI) and (2) snowball sampling strategies where participating mothers shared study information by word of mouth or social media with other mothers that might be eligible. The PI (first author) also virtually attended the community organizations’ monthly support group meetings and shared study information with other local support groups’ social media and email lists.

The children with DDs were included based on the federal definition found in the Developmental Disabilities Assistance and Bill of Rights Act of 2000. Inclusion criteria were mothers who were (1) self-identified mothers and primary caregivers of school-aged (ages 6–12) children with DDs, including autism spectrum disorder (ASD), cerebral palsy, genetic disorders, such as Down syndrome and Fragile X syndrome, and other developmental disabilities; (2) 21 years of age or older; and (3) able to read, speak and write English. Exclusion criteria included mothers who were: (1) pregnant (due to changes in sleep patterns); (2) a shift worker (changes in sleep patterns); (3) had a history of a sleep disorder; and (4) diagnosed with severe mental health conditions.

Measures and Variables

Feasibility Measures

Feasibility outcomes collected included recruitment rate, adherence, and acceptability. Adherence was defined as the following two criteria are met: (1) mother’s completion of at least five of seven consecutive days/nights wearing actigraphy watches and (2) children with DDs completing a minimum of five of seven consecutive nights recording videosomnography. Acceptability to using actigraphy and videosomnogrsphy was assessed through an adopted exit survey about study participation. The eight-item survey was answered on a 5-point Likert scale with responses ranging from one (strongly disagree) to five (strongly agree). Items addressed enjoyability of study participation, ease of use with wearing equipment, and willingness to participate in future research (e.g., “my time as a participant has been enjoyable,” “the video camera was easy to set up,” “I would participate in this study again”). Open-ended questions also were included to obtain mothers’ likes and dislike about participation.

Sleep-wake Variables

Definitions of sleep-wake variables from actigraphy and videosomnography are displayed in Table 1.

Table 1 Sleep-wake variables from actigraphy and videosomnography
Full size table

Mothers’ Sleep Measurement

Mothers’ sleep-wake variables were obtained using the Actiwatch-2 (Philips Respironics, Bend, Oregon, U.S.). The device collected data in 30-second epochs with medium sensitivity (wake threshold value = 40 activity counts for one minute epochs). Mothers wore the actigraphy watch on their non-dominant wrist for seven consecutive days/nights, except during showering and swimming. We downloaded actigraphy data from the watch to the Actiware software (version 6.1.2) and manually reviewed it based on the decision rules adopted from published scoring procedures (Adams et al., 2019; Ancoli-Israel et al., 2015; Chow et al., 2016; Fekedulegn et al., 2020; Li et al., 2021; Meltzer & Westin, 2011). Actigraphy had good reliability and validity in differentiating sleep from wakefulness (Sadeh & Acebo, 2002) and comparable sleep-wake parameters to polysomnography (Jean-Louis et al., 2001).

We defined sleep onset/offset times using the following criteria: (1) sleep onset time was defined as the first minute of at least three consecutive minutes when there was a drop in activity level and decreased light level to zero and/or the event marker pressed; and (2) the sleep offset time was defined as the last minute of five consecutive minutes when there was an increase in activity level, a solid wake bar (red bar), increased light level above 1.0 µW/cm2, and/or event marker pressed. WASO was calculated between sleep onset/offset time based on (1) the event marker or (2) the solid wake bar (red bar) showed activity for more than two minutes without the event marker. The daytime naps were defined based on no movement for a minimum duration of 30 min outside the sleep onset/offset time compared with the mother’s sleep dairy-reported naps. If naps were within 60 min from sleep onset/offset time or the segment that lasted longer than three consecutive hours with almost zero activity counts, we did not score as naps.

Mothers’ perceived sleep quality was measured using the well-established Pittsburgh Sleep Quality Index (PSQI) (Buysse et al., 1989). The PSQI is a 19-item self-rated measure of sleep quality over the past month. The PSQI has seven component scores on a zero (no difficulty) to three (severe difficulty) scale, which was summed for a global score ranging from zero to 21, with scores greater than five indicating poor sleep quality (Buysse et al., 1989). The PSQI demonstrated good test-retest reliability (α = 0.87) and good convergent validity in patients with primary insomnia (Backhaus et al., 2002). In our study, Cronbach’s α for the total scale was 0.81.

Child’s Sleep Measurement

The child’s sleep-wake variables and the mother’s nighttime caregiving activities were measured using videosomnography. Videosomnography is a portable time-lapse video recording system that has been used to collect sleep-wake patterns and other nighttime behaviors, such as mothers’ caregiving activities (Ipsiroglu et al., 2015). A video recording device with storage and a night-vision camera (CammPro I826 Camera) was placed in the child’s primary sleep location for seven consecutive nights. We analyzed video recordings at 1-minute intervals. Two trained coders coded videos using a structured guide for specific behaviors. Videosomnography has been used to measure sleep behaviors in children with ASD (Moore et al., 2017) and validated with actigraphy in sleep-wake patterns in infant and toddler sleep (Schwichtenberg et al., 2018; Sitnick et al., 2008).

A sleep diary was used to aid in interpreting actigraphy and videosomnography data to clarify mothers’ and children’s sleep onset/offset time. We used an adapted version from the Consensus Sleep Diary (CSD) (Carney et al., 2012). The CSD has items about the quantitative dimensions of sleep (e.g., TST, sleep disruptions) and sleep-related factors (e.g., alcohol, caffeine, medication use). We modified the CSD by adding items focusing on children with DDs’ sleep (e.g., time to put a child to sleep, child’s sleep offset time, # of wakings due to a child’s care) and a brief description of any family events that influenced sleep.

Additional Variables are Known to be Associated with Mothers’ Sleep

Mothers’ depressive symptoms were measured by the Center for Epidemiologic Studies Depression Scale (CES-D) (Radloff, 1977), a 20-item self-report of the frequency of depressive symptoms during the past week on a zero to three scale with total scores from zero to 60. Higher scores indicate higher depressive symptoms, and cut-off scores of ≥ 16 indicate possible depression. The CES-D showed adequate test-retest reliability (r = .71), and construct validity (Cosco et al., 2017). Our study’s Cronbach’s α for the total scale was 0.93.

Caregiver stress was measured by an adapted short version (with permission) of the Zarit Burden Interview (ZBI-12) (Bédard et al., 2001). The Parenting Stress Index (PSI) has been widely used to measure parenting stress, but it was not specifically designed for parents of children with DDs. The factor analysis failed to support two subscales of the PSI in parents of children with DDs (Dardas & Ahamd, 2013; Zaidman-Zait et al., 2011). While ZBI was originally developed for caregivers of persons with Alzheimer’s or dementia, the modified version of the ZBI has been used in parents of children with DDs and validated by other caregiver populations (Bachner et al., 2007). The ZBI-12 is a 12-item questionnaire measuring strain associated with the caregiving of persons with dementia. Questions from the ZBI-12 were modified by replacing “your relative” with “your child. The responses ranged from never (0) to nearly always (4), with possible scores of zero to 48. Higher total scores reflect greater caregiving stress, with ≥ 17 as the cut-off point for high caregiving burden (Bédard et al., 2001). The ZBI-12 has demonstrated good reliability (α = .88) and convergent validity in caregivers of patients with dementia (Bédard, 2001). In our study, Cronbach’s α for the total scale was .75.

Child’s sleep problems were subjectively measured by mothers’ reports on their child’s sleep over the previous week using the abbreviated Children’s Sleep Habit Questionnaire (CSHQ) (Owens et al., 2000). The CSHQ is a 33-item self-report questionnaire that was developed as a screening tool for sleep problems among school-aged children. The CSHQ consists of eight subscales: Bedtime Resistance, Sleep Onset Delay, Sleep Duration, Sleep Anxiety, Night Waking, Parasomnias, Sleep Disordered Breathing, and Daytime Sleepiness. Items are rated on a 3-point scale as occurring “usually,” “sometimes,” and “rarely.” The total CSHQ scores ranged from 33 to 99, and higher scores indicate greater sleep problems, with ≥ 41 as the clinical cut-off score. The original CSHQ has been used in studies with children with DDs (Mannion et al., 2013; Stuttard et al., 2015). In our study, three subscales (Sleep Anxiety, Nighttime Waking, and Parasomnia) had low internal consistency reliability coefficients. We systematically removed items with negatively corrected item-to-total correlations to obtain acceptable alpha coefficients. This resulted in two items in the Sleep Anxiety (α = 0.90), two items in the Night Waking (α = 0.71), and three items in the Parasomnias (α = 0.72) subscales. The items removed are reported as single items (See Table 2). The Cronbach’s α coefficient for subscale scores ranged from 0.67 to 0.90.

Table 2 Sample characteristics and responses to exit survey (n = 10)
Full size table

We developed a questionnaire to collect information on mothers’ and child’s demographic and health information. The questionnaire included items for mothers’ age, race, work status, health problem, other children, and their ages. Child items addressed types of disabilities, additional medical conditions, caregiving needs, and therapies received.

Procedures

We obtained Institutional Review Board approval and written informed consent from all mothers and verbal or written assent from all children with DDs. Due to the COVID-19 pandemic, we developed and implemented alternative data collection strategies, such as the advanced mailing of the consent and questionnaires, meeting participants via a virtual meeting, and creating video-recorded instructions on recording a child’s sleep.

Initial Contact and Baseline Virtual Visit

The PI contacted interested mothers for an initial telephone screening to confirm their willingness to participate, assess eligibility, and schedule a virtual meeting. After the initial screening, the PI mailed the informed consent and assent forms and the questionnaires (PSQI, CES-D, ZBI, and CSHQ) to the participant’s home.

The mother and the child with DDs were at a secure video virtual meeting. During the virtual meeting, the PI obtained consent from the mothers. Depending on the child’s age and ability, assent was obtained. The mother’s and child’s demographic information and medical history also were obtained. The PI scheduled a home visit to drop off the equipment. After the virtual meeting, mothers received a text or email with the link for a short video-recorded instruction on setting up and recording the child’s sleep. Mothers were encouraged to visit the link before the home visit.

Home Visits

The PI or trained research assistant (RA) delivered the package to the participant’s home, adhering to the Centers for Disease Control and Prevention’s infection control guidelines (Centers for Disease Control and Prevention [CDC], 2022). The package contained the actigraphy and videosomnography equipment, written instructions about wearing the actigraphy watch and video recording, the 7-day sleep diary, and the exit survey. During the home visit, the PI or RA briefly reviewed the equipment and written instructions with each mother. The mother was encouraged to demonstrate how to record with the camera and ask questions. At the end of this home visit, signed consent/assent forms and completed questionnaires were collected.

The second home visit was scheduled shortly after the week of data collection. The PI or RA picked up the equipment, sleep diary, and exit survey outside the participant’s home. The mothers received participant remuneration (a $100 gift card to the local store).

Actigraphy and Videosomnography Protocols

Mothers were instructed to press the watch event marker for the sleep onset/offset, any wakings, and onset/offset of a nap. Mothers also were instructed on using the daily sleep diary. Mothers were asked to set up the camera at the foot of the child’s bed at least 40 inches above the bed and start and end video recording their child’s sleep each night. After the first night of recording, the PI sent text messages to identify any recording-related issues. Two other text messages were sent on the third and fifth days to address any issues. The pick-up time was scheduled in the last text message, and mothers were reminded to complete the exit survey.

Data Analysis

Data were analyzed using SPSS Version 27.0 and SAS Version 9.3. The primary products of this work pertain to feasibility, and thus our study was not powered to conduct hypothesis testing. Mothers and their children with DDs’ sleep-wake variables were averaged across nights for each participant and reported with appropriate central tendency measures (mean or median) and dispersion (standard deviation or interquartile range). Sleep onset/offset time and total sleep time across the seven days were presented graphically (Figs. 1, 2, 3 and 4). Descriptive statistics were used to summarize sample characteristics, responses to standard questionnaires, sleep data, and exit survey items. Rates of recruitment, adherence, and acceptability were quantified. The frequency of missing data was also described.

Fig. 1
figure 1

Mothers’ Sleep Onset Time and Sleep Offset Time by Night from 7-Night Actigraphy

Hours are presented by hh:mm on a 24-hour clock

Full size image
Fig. 2
figure 2

Mothers’ Total Sleep Time in Minutes from 7-Night Actigraphy

Full size image
Fig. 3
figure 3

Child’s Sleep Onset Time and Sleep Offset Time from 7-night Videosomnography

Hours are presented by hh:mm on a 24-hour clock

Full size image
Fig. 4
figure 4

Child’s Wake After Sleep Onset in Minutes from 7-Night Videosomnography

Full size image

Results

Sample Characteristics

Mothers and their children with DDs’ characteristics are presented in Table 3. The mothers’ mean age was 41.3 (SD = 5.7). Most mothers were African Americans, married or living with a partner, working full-time or part-time, and had at least a college degree. Six out of 10 mothers reported having one or more health conditions, including hypertension, type 2 diabetes, back pain, or a history of depression and anxiety. The children with DDs were 10 (SD = 1.56) years old on average. Half of the children were boys and had ASD. In addition to the primary diagnosis, children with DDs had at least one or more health conditions. The two most common conditions were epilepsy and attention-deficit/hyperactivity disorder (ADHD). Three children with DDs were taking clonidine for sleep. Children also were taking antiepileptic (five children), ADHD medicine (three children), antipsychotics (two children), and muscle relaxants (one child) that may have influenced sleep.

Table 3 Sleep-wake variables from actigraphy and videosomnography and sleep-related variables
Full size table

Recruitment Rate

Across four months during the height of the COVID-19 pandemic (late September 2020 to early February 2021), 33 mothers were contacted; 10 were lost to follow-up before screening; nine did not meet screening criteria, and one was excluded because the mother had an infant in addition to the child with DDs, which may influence mother’s sleep patterns. Of the 13 mothers and children contacted and eligible, three declined to participate after screening resulting in 10 enrolled in the study (76.9% recruitment rate).

Adherence

Ten mothers wore actigraphy watches for seven consecutive days/nights; however, we could not retrieve data from two watches because of technical issues. Of the eight mothers with retrievable actigraphy data, the adherence rate for at least five consecutive days/nights was 100%, all having usable data for seven nights. For child’s sleep, nine mothers recorded their child’s sleep on average 5.9 (SD = 2.3) nights, ranging from four to seven nights, with an adherence rate for at least five consecutive nights of 89%.

Acceptability

Overall, mothers reported study participation positively and using the equipment as acceptable. The mean rating of eight questions was 4.6 (SD = 0.4)(out of five). Table 3 includes the responses to structured items from the exit survey. In the open-ended questions, mothers indicated that what they liked the most about the study was “[The study equipment was] easy to use. The questions were simple to understand and complete. It did not take a lot of time”; “I liked being able to track [my] sleep as well as my daughter’s [sleep]. I also noticed my activities through[out] the day”; “It helped me focus more on myself and my routine for my family”; and “The frequent communication and set up instructions made it easy.” When asked what the mothers liked the least about the study, several comments were related to wearing an actigraphy watch: “The watch was awkward, and I kept forgetting to press the button”; “The watch. It took me one to two days to get used to it. I would forget to push the button to mark sleep/wake at times”; and “I had to wear my smartwatch on my dominant wrist because actiwatch kept smacking it, which took some time to get used to!” A few comments were related to videosomnography: “Camera didn’t stay attached [on the plate of the tripod]. I had to put a strap to hold it”; and “wish the camera would stick to the wall…unprepared [for] moments that involved my daughter’s pullup showing on screen due to tossing & turning& waking up early.”

Missing Data

While we obtained 7-night sleep data from 8 mothers, we could not retrieve data from two watches due to malfunction that was unable to be recovered by the actigraphy manufacturer or us. This resulted in a 20% missing data rate for nights with actigraphy. Nine out of 10 mothers completed video recordings of their child’s sleep with a missing data rate of 10% for videosomnography. Two children had missing data that ranged between one and three nights.

Sleep-Wake Variables

Mothers’ Sleep Measurement

A summary of mothers’ sleep-wake variables from actigraphy is presented in Table 4. Mothers’ median sleep hours were within the recommended amount of sleep (≥ 7 h per night) for a healthy adult (Watson et al., 2015). On average, mothers woke up 3.1 times (SD = 0.9) nightly with a median WASO of 10 min (5.0, 19.3). Three mothers took a nap at least once during the week (one, four, and five times respectively). Sleep efficiency was within the standard limit for adults (≥ 80%) (Berger et al., 2005). High variability in sleep timing (sleep onset/offset time), as well as sleep consistency (TST over seven night period) within and between mothers, were noted (See Figs. 1 and 2). The mean PSQI score was 6.5 (SD = 3.4), and five of eight mothers who completed the PSQI had scores above the cut-off point (> 5), indicating poor sleep quality.

Table 4 Descriptive statistics for child sleep problems (Children’s Sleep Habit Questionnaire, n = 10)
Full size table

Child’s Sleep Measurement

Table 4 also includes the summary of children with DD’s sleep-wake variables from videosomnography. About 13% of the child’s sleep onset times (one sleep offset time from one child) were not captured (mainly from the first three participants), and we used the sleep diary for those times. On average, children with DDs slept 6.4 (SD = 1.9) hours, substantially less than the recommended 9–12 h for school-aged children to promote optimal health (Paruthi et al., 2016). Children with DDs woke up on average 3.4 (SD = 1.6) times nightly with WASO of 79 (16.0, 183.0) minutes. While recommended children’s bedtime is no later than 9:00 PM (Mindell et al., 2009), only one child’s average sleep onset time fell within this recommendation. Children had high variability in sleep timing over the seven nights (Fig. 3), especially for WASO time (Fig. 4).

Mothers came to a child’s bedroom for various reasons, and they included (1) adjustment of the child’s sleep environment (e.g., readjusting covers/child’s position); (2) providing care for a child, including changing a diaper and giving medication or water; and (3) turning off the musical toys, lights or a TV. Noticeably, five out of nine children were co-sleeping with either their mothers or someone else (e.g., siblings) for at least two nights over the 7-night period (ranging from two to five nights).

Additional Variables are Known to be Associated with Mothers’ Sleep

The average depressive symptoms score was close to the cut-off score of 16 (mean [M] = 15.1, SD = 10.8), but the PSQI scores were highly variable (ranging from three to 34) (Table 4). The average caregiving (ZBI) score was high (M = 19.6, SD = 7.0), and six of 10 mothers reported total scores above the cut-off scores (≥ 17), indicating a high caregiver burden. From sleep diaries, mothers reported waking up on average 35.4 (SD = 35.4) minutes earlier than planned. Mothers also reported specific events that may affect their sleep, and they included (1) using the bathroom, (2) having a monthly cycle, (3) feeling down or not well, and (4) the child’s school schedule (e.g., out-of-school or staying home due to remote learning). Table 2 presents CSHQ subscale scores and individual items. The average CSHQ score was 57.3 (SD = 6.7), and all children with DD’s CSHQ total scores were above the clinical cut-off point (≥ 41). Mothers rated their child’s sleep problems high in Bedtime Resistance, Sleep Onset Delay, and Daytime Sleepiness subscales.

Discussion

We found using actigraphy and videosomnography was feasible and acceptable for measuring sleep-wake variables for mothers and their school-aged children with DDs. We obtained sleep-wake variables for seven consecutive nights, a sleep diary, and self-report instruments with minimal missing data. In addition, most mothers felt that using actigraphy and videosomnography was easy and indicated a willingness to participate in the study again.

Use of Actigraphy and Videosomnography in Mothers and Their Children with DDs

Mothers noted that frequent communication using text messages across seven days was helpful to facilitate adherence to the actigraphy and videosomnography. While most mothers reported that using actigraphy was easy, some commented that pressing the watch event marker was difficult because of the lack of feedback or a sensor on the device’s face. The mothers suggested a light or sound after the watch event marker is pressed would be helpful. A couple of mothers asked about using commercially available wearable sleep devices. While wearable sleep devices continue to advance and are less costly, research-grade actigraphy watches (such as Actiwatch-2) provide better sleep-wake estimation than commercially available wearable devices (Evenson et al., 2015; Van den Water et al., 2011).

While missing data from actigraphy and videosomnography were minimal, additional steps can be used to minimize it in future studies. Although each watch was fully charged and initialized before each use, technological issues occurred, resulting in data loss from the two watches. Testing the watch by wearing it for an extended period may identify issues and prevent data loss. We could not capture some children’s sleep onset time from videos, especially from earlier participants. The sleep diaries indicated that some children fell asleep outside their bedrooms. Another explanation may be that video recording instructions were unclear. We noted this issue early and changed the protocol to include verbal reinforcement of video recording time during the first home visit. Subsequently, mothers improved capturing the child’s sleep onset time. Future studies may consider adopting these protocols to minimize missing data.

Contributions to Future Research and Practice

Consistent with previous research (Lee et al., 2018a; Micsinszki et al., 2018), mothers reported poor sleep quality in our study. However, mothers’ sleep quantity was within the recommended sleep hours. Our previous study found that 40% of the mothers of school-aged children with DDs slept less than recommended hours nightly (Lee et al., 2018b); however, sleep measurement was self-report using a 5-day sleep diary. This discordance between mothers’ self-reported and objective sleep measures may be due to increased stress and depressive symptoms. In an early study, stress and depressive symptoms explained the differences in subjective and objective sleep measures (Maglione et al., 2012). We cannot examine these relationships; however, mothers reported high caregiver stress, and several had high depressive symptoms, which may influence mothers’ perception of sleep quality. Significant associations between sleep quality and parenting stress and depressive symptoms have been reported in mothers of children with DDs (Gallagher et al., 2009b; Lee, 2013).

Our study illustrates that simultaneous examination of the mother’s and child’s sleep using objective sleep measures over several nights can provide a comprehensive view of what might contribute to the mother’s poor sleep quality, including nighttime caregiving activities, and a better understanding of how the child’s sleep problems affect mothers’ sleep. The figures show that mothers of children with DDs slept late and showed high variability in sleep timing and consistency. We also noted various nighttime caregiving activities via videosomnography. If a child’s lack of sleep at night is associated with more daytime behavior problems, this can increase the mother’s stress levels, which may impact the mother’s perceived sleep quality. A significant association between sleep problems and children with DDs’ behavior problems (Esbensen et al., 2018) and a transactional relationship between a child’s behavior problems and maternal stress have been reported. (Woodman et al., 2015). Both child’s sleep and behavior problems were significantly related to maternal stress in mothers of children with autism and other DDs (Valicenti-McDermott et al., 2019). We did not include a measure of child behavior problems, which may be important to include in future research.

Children with DDs in our study had inadequate sleep hours. We noted the lights were on and noises outside of the child’s bedroom, and children used a tablet or watched TV while awake. The figures indicate the child’s awakening variability at night and some for extended periods. However, from the videos, we found that many of child’s awakenings had no parental interventions. Mothers may be unaware of how little children with DDs are sleeping. Giving mothers information about their child’s sleep behaviors so mothers can change the sleep environment may improve their children’s sleep. Although mothers were not awake all of the time that their children were during the night, if a child routinely has prolonged nightly awakenings, these are potential times the mother may become involved.

Mothers of children with DDs also may benefit from having a tailored sleep health intervention addresses mother- and child-related factors affecting mothers’ sleep. Mothers’ sleep quality can be improved by improving sleep timing and minimizing awakenings at night. A few intervention studies focusing on stress reduction and mindfulness lowered stress and depression in parents of children with DDs (Dykens et al., 2014; Neece et al., 2013). Cognitive and behavioral sleep health interventions that included stress management sleep hygiene, and exercise components showed a medium effect on improving sleep quality in adults without clinical sleep disorders (Murawski et al., 2018). This study found that children with DDs slept significantly less than recommended hours. Sleep problems are prevalent in children with DDs and affect mothers’ sleep (Bourke-Taylor et al., 2013; Valicenti-McDermott et al., 2019). Poor sleep quality was a significant predictor of both mothers’ mental and physical health (Lee et al., 2018a) and mediated the association between heart rate variability (as a proxy of chronic stress) and depressive symptoms (Estrela et al., 2021) in mothers of children with DDs. Later sleep timing and higher sleep variability in sleep duration were associated with adverse mental and physical health outcomes (Chaput et al., 2020). Future research may need to examine sleep and mental and physical health over time.

Limitations and Strengths

This study also has limitations. First, the non-random and small sample affects generalizability. Second, caution is needed in interpreting the total CSHQ scores and scores in some subscales due to low internal consistency reliability.

Despite these limitations, we believe our study is the first to use videosomnography to measure sleep-wake variables for school-aged children with DDs. Two studies have reported using videosomnography for infants/toddlers with DDs, including ASD (Ipsiroglu et al., 2015; Schwichtenberg et al., 2018), but none for school-aged children with DDs. Using objective sleep measurements, we simultaneously captured the mothers’ and child’s sleep-wake variables, including child-related nighttime caregiving activities, which helped determine if future sleep interventions for these mothers need to focus on improving nighttime caregiving activities. We also successfully recruited vulnerable populations during the COVID-19 pandemic and showed that using these objective sleep measures was feasible.

Conclusion

Sleep is essential for good physical and mental health. We demonstrated that actigraphy and videosomnography were feasible for objective measurement in mothers and their children with DDs. These mothers and children had highly variable sleep patterns, and children awake during the night for a substantial amount of time. Finding ways to improve the sleep of mothers and children with DDs will likely result in positive family health outcomes.