- Journal List
- HHS Author Manuscripts
- PMC6033696
As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsem*nt of, or agreement with, the contents by NLM or the National Institutes of Health.
Learn more: PMC Disclaimer | PMC Copyright Notice
J Abnorm Child Psychol. Author manuscript; available in PMC 2019 Aug 1.
Published in final edited form as:
Cara A. Palmer, Ph.D.,1 Michelle A. Clementi, M.A.,1 Jessica M. Meers, M.A.,1 and Candice A. Alfano, Ph.D.1
Author information Copyright and License information PMC Disclaimer
The publisher's final edited version of this article is available at J Abnorm Child Psychol
Abstract
Little is known about the co-sleeping behaviors of school-aged children,particularly among anxious youth who commonly present for the treatment of sleepproblems. The current study examined the occurrence of co-sleeping in bothhealthy and clinically anxious children and its associated sleep patterns. Atotal of 113 children (ages 6–12), 75 with primary generalized anxietydisorder and 38 healthy controls, participated along with their primarycaregiver. Families completed structured diagnostic assessments, and parentsreported on their child’s co-sleeping behaviors and anxiety severity.Children provided reports of anxiety severity and completed one week ofwrist-based actigraphy to assess objective sleep patterns. A significantlygreater proportion of anxious youth compared to healthy children co-slept, andgreater anxiety severity was related to more frequent co-sleeping. Co-sleepingin anxious youth was associated with a delay in sleep timing and with greatersleep variability (i.e., more variable nightly sleep duration). All analysescontrolled for child age, race/ethnicity, family income, and parental maritalstatus. Co-sleeping is highly common in anxious school-aged children, with morethan 1 in 3 found to co-sleep at least sometimes (2–4 times a week).Co-sleeping was even more common for youth with greater anxiety severity.Increased dependence on others to initiate and maintain sleep may contribute topoorer sleep in this population via shifted schedules and more variable sleeppatterns.
Keywords: anxiety disorders, co-sleeping, sleep dependence, sleep variability
The term ‘co-sleeping’ is broadly used to refer to childrenwho sleep in close proximity to parents and often includes bed sharing (sharing asleeping surface with a family member for all or a portion of the sleep period;Goldberg and Keller 2007; McKenna and Volpe 2007). Co-sleeping/bedsharing is a controversial topic in the pediatric sleep literature, with no generalconsensus regarding its potential benefits or consequences (Thoman 2006). To a large extent, co-sleeping is a culturalphenomenon, widely accepted in Eastern societies that prioritize collectiveness anddependency, and less practiced in Western societies where autonomy is emphasized(Latz et al. 1999; Mileva-Seitz et al. 2017). Across nearly all cultures however,co-sleeping is most common in infancy and early childhood and declines as childrenenter school (BaHammam et al. 2008; Cortesi et al. 2004; Jenni et al. 2005; Li et al. 2009). Still, reported rates ofco-sleeping among children in the United States vary greatly (anywhere from5–88%), with very little understanding of the factors that influenceor sleep patterns that result from this practice, particularly in school-agedchildren (Cortesi et al., 2004; Jenni et al., 2005; Latz et al. 1999; Mileva-Seitzet al. 2017; Welles-Nystrom, 2005).
A need to more closely examine co-sleeping behaviors in school-aged childrenis derived from evidence linking co-sleeping in middle to late childhood withgreater levels of sleep-related difficulties and anxiety/fear (Blader et al. 1997; Cortesi etal. 2004; Liu et al. 2003; Mindell et al. 2009). For example, co-sleepingduring the school-aged years has been associated with problems initiating sleep,less nighttime sleep, more daytime sleepiness, more bedtime resistance, increasednighttime awakenings, and greater levels of sleep anxiety (Blader et al. 1997; Cortesi etal. 2008; Jiang et al. 2016).Investigations of co-sleeping are especially needed among anxious children, whooften experience heightened fear/arousal surrounding sleep as well as problemsinitiating sleep, which can prompt requests to co-sleep with parents or siblings(Alfano et al. 2007; Alfano et al. 2010; Forbes etal. 2006; Hansen et al. 2011).Children with anxiety disorders are commonly referred for problems sleepingindependently (Beidel and Alfano 2011; Cowie et al. 2014) and capitulation toco-sleeping requests (i.e., as a means of preventing or reducing child distress)ranks among the most common and most interfering forms of parental accommodationamong school-aged children with anxiety disorders (Thompson-Hollands et al. 2014). Paradoxically, however, investigationson the occurrence and associated outcomes of co-sleeping in clinically-anxious youthare virtually nonexistent in the literature. The aim of the current study was tofill this research gap by examining rates of co-sleeping and relationships betweenco-sleeping and objective sleep patterns in a sample of clinically-anxious andnon-anxious school-aged children.
Co-Sleeping and Anxiety in School-Aged Children
Children’s requests to co-sleep are often reactive or occur inresponse to previous sleep difficulties or anxieties (Cortesi et al. 2004; Cortesi et al. 2008; Jain et al.2011; Rath and Okum 1995;Smedje et al. 2001). For example,children who struggle to self-soothe at night have more difficulty fallingasleep independently and often ‘signal’ their distress toparents (Sadeh and Anders 1993). Theperiod leading up to sleep is particularly stressful for clinically-anxiouschildren, who experience higher subjective levels of cognitive and physiologicalarousal (Alfano et al. 2010), highercortisol levels (Forbes et al. 2006), andgreater anxiety and fear (Hansen et al.2011) during the pre-sleep period than age-matched peers. Togetherwith difficulties regulating distress more broadly (Carthy et al. 2010; Suvegand Zeman 2004), anxious youth might be more likely to signal thisdistress to parents, giving rise to parental accommodation in the form ofco-sleeping (Thompson-Hollands et al.2014).
Preliminary evidence suggests co-sleeping is prevalent among school-agedchildren with anxiety disorders (Alfano et al.2007; Storch et al. 2007).Among one study investigating 6–11 year-old children with variousanxiety disorders, 61% of parents reported that their child commonlyresisted or refused to sleep alone (Alfano et al.2007). However, in addition to the lack of a healthy comparison groupof children, the actual occurrence of co-sleeping was not the focus of thisinvestigation, and it remains unknown whether co-sleeping in school-agedchildren with anxiety disorders relates to both anxiety severity and/orproblematic sleep patterns.
Co-Sleeping and Child Sleep Patterns
Having a parent present during sleep onset is one of the strongestpredictors of poor sleep patterns throughout childhood (Mindell et al. 2009). Although typically limited to onlysubjective parent-reported sleep disturbances, numerous cross-sectional studiesin school-aged children find co-sleeping to be a risk factor for sleep problems.For example, co-sleeping has been related to greater night awakenings (Cortesi et al. 2004; Keller and Goldberg 2004; Mindell et al. 2009), shorter sleep duration (Cortesi et al. 2008), more bedtime resistance (Cortesi et al. 2008; Jiang et al. 2016), poor sleep quality (Jiang et al. 2016) and more daytimesleepiness (Jiang et al. 2016; Liu et al. 2003).
Conceivably, relationships between co-sleeping and sleep problems resultfrom alterations or inconsistency in sleeping arrangements and schedules (Lozoff et al. 1984; Peterman et al. 2015). Because co-sleeping/bed sharinginherently introduces greater variability into the sleep routine and creates adependence on the schedule of other family members, both bedtimes and awakeningtimes may shift (Li et al. 2009). Cross-sectional research has indeed foundchildren who share a bedroom experience greater intraindividual variability insleep onset and offset times (Buckhalt et al.2007), and co-sleeping children are more likely to go to bed later(Cortesi et al. 2008) and have lessregular bedtimes (Hayes et al. 2001).
A host of research emphasizes organization of the sleep-wake system inhealthy developmental outcomes, including the timing and regularity of sleeppatterns. A greater degree of instability and irregularity has been found topredict worse behavioral and health problems (Thelen 1993; Thoman 1990).Independent of sleep quantity, inconsistent sleep schedules in children areassociated with poor mental health, daytime functioning, and school performance(Becker et al. in press). Further, ashifted sleep-wake cycle (e.g., a later bedtime) in youth is related to greateremotional distress or poorer health outcomes both concurrently and later inadulthood (McGlinchey and Harvey 2015),including greater depressive symptoms (Emens etal. 2009; Short et al.2013).
Sleep Timing and Variability among Anxious Children
Subjective reports of sleep problems among anxious youth are highlycommon (Alfano et al. 2007; Alfano et al. 2010). However, consistencyand/or timing of sleep and their impact on daytime functioning have received farless attention among children with anxiety disorders (Alfano et al. 2007; Cowie etal. 2014; Hansen et al. 2011;Peterman et al. 2015). One studyfound clinically-anxious youth to go to bed significantly later than healthycontrols on school nights (based on sleep diary reports; Hudson et al. 2009). The same study found significantvariability in bedtimes on weekdays and weekends among both clinically-anxiousand healthy youth (Hudson et al. 2009).The fact that only subjective sleep reports were used, which do not alwayscorrespond with objective measurement of sleep, particularly for clinicallyanxious youth, to some extent limit these findings (Alfano et al. 2015). However, variability in sleep timing isalso concurrently associated with increased anxiety symptoms in healthyschool-aged children (Fletcher et al. inpress) and research in adolescents has linked anxiety symptoms andvariability in sleep time (Fuligni and Hardway2006) independent of mean sleep duration. Notably, none of thesestudies considered the potential role of co-sleeping behavior.
Current Study
The current study sought to investigate the occurrence of co-sleepingand its correlates in a school-aged sample (6–12 years old) ofclinically-anxious and non-anxious children. Our first aim was to document andcompare the frequency of parent-reported co-sleeping in these two groups ofchildren, and to examine associations with anxiety severity within the clinicalsample (based on parent, child, and clinician ratings). We hypothesized thatanxious youth would co-sleep significantly more often than non-anxious youth,and that more frequent co-sleeping would be related to greater levels ofanxiety.
Based on previous research suggesting that both increased anxiety andco-sleeping relate to greater sleep problems (Mindell et al. 2009) and to later shifts and greater variability insleep patterns (Buckhalt et al. 2007;Doane et al. 2015; Fuligni and Hardway 2006; Hudson et al. 2009; Li et al. 2009), thesecond aim of our study was to examine relationships among co-sleeping and sleeppatterns within a clinically-anxious sample. Building on some of the limitationsof previous studies, we examined these associations using an objective sleepmeasure (actigraphy). Specifically, we examined relationships betweenco-sleeping and mean total sleep duration, night wakings (number of awakenings,time spent awake after the onset of sleep), and sleep timing (sleep onset andoffset times). Additionally, we examined relationships among co-sleeping andintraindividual variability in sleep (i.e., day-to-day variability in sleepduration and sleep timing). Consistent with prior findings, we hypothesized agreater frequency of co-sleeping to be associated with more problematic sleeppatterns.
Importantly, because decisions to co-sleep may be pragmatic or culturalin nature (Owens 2004), and becauseco-sleeping tends to decline with age (Mindellet al. 2009), all analyses controlled for age, race/ethnicity,income, and parent marital status.
Method
Participants
Participants included 113 youth ages 6–12 and a primarycaregiver. Demographic characteristics are reported in Table 1. Children were diagnosed with primary generalizedanxiety disorder (GAD; n = 75) or were healthy controls(n = 38). Three additional families participated, but didnot complete the co-sleeping measure. These participants were not included inthe current study. Children were required to be fluent in English, live withtheir parent/caregiver for minimum of one year prior to assessment, and beenrolled in regular education classes. Any children with a current or lifetimehistory of psychotic, pervasive developmental, bipolar, eating, or substance usedisorder, or suicidal ideation/self-harm were excluded from participation. Otherexclusion criteria included the child’s use of psychiatric medicationsor medications known to impact sleep, any chronic illness that could impactsleep, or current use of treatment services for anxiety or sleep problems.
Table 1
Group Differences in Demographics and Variables of Interest
| Anxious | Control | Total Sample | Comparison Statistics | |||
|---|---|---|---|---|---|---|
| M (SD) / % (n) | M (SD) / % (n) | M (SD) / % (n) | t or X2 (df) | p | ||
| Age | 9.01 (1.70) | 8.92 (1.26) | 8.98 (1.56) | −.33 (95.77) | .75 | |
| Gender (Female) | 50.7% (38) | 52.6% (20) | 51.3% (58) | .04 (1) | .84 | |
| Race/Ethnicity | 3.00 (4) | .56 | ||||
| Caucasian | 54.7% (41) | 60.5% (23) | 56.6% (64) | |||
| African-American | 4% (3) | 5.3% (2) | 4.4% (5) | |||
| Asian | 1.3% (1) | 2.6% (1) | 1.8% (2) | |||
| Hispanic | 13.3% (10) | 18.4% (7) | 15% (17) | |||
| Other/Biracial | 26.7% (20) | 13.2% (5) | 22.1% (25) | |||
| Income | −1.58 (109) | .12 | ||||
| < $10K | N/A | 5.4% (2) | 1.8% (2) | |||
| $10–20K | 2.7% (2) | 2.6% (1) | 2.7% (3) | |||
| $20–40K | 8.1% (6) | 10.8% (4) | 9% (10) | |||
| $40–60K | 8.1% (6) | 5.4% (2) | 7.2% (8) | |||
| $60–80K | 10.8% (8) | 16.2% (6) | 12.6% (14) | |||
| $80–100K | 10.8% (8) | 16.2% (6) | 12.6% (14) | |||
| >$100K | 59.5% (44) | 43.2% (16) | 54.1% (60) | |||
| Maternal Education | −1.38 (110) | .17 | ||||
| Some grade school | 1.4% (1) | 2.6% (1) | 1.8% (2) | |||
| Completed high school | 1.4% (1) | 10.5% (4) | 4.5% (5) | |||
| Some College | 17.6% (13) | 10.5% (4) | 15.2% (17) | |||
| Completed College | 40.5% (30) | 47.4% (18) | 42.9% (48) | |||
| Advanced Degree | 39.2% (29) | 28.9% (11) | 35.7% (40) | |||
| Paternal Education | −.69 (4) | .50 | ||||
| Some grade school | 1.4% (1) | N/A | .9% (1) | |||
| Completed high school | 11% (8) | 5.6% (2) | 9.2% (10) | |||
| Some College | 17.8% (13) | 36.1% (13) | 23.9% (26) | |||
| Completed College | 34.2% (25) | 33.3% (12) | 33.9% (37) | |||
| Advanced Degree | 35.6% (26) | 25% (9) | 32% (35) | |||
| Parent Marital Status | 1.25 (1) | .26 | ||||
| Married | 81.3% (61) | 89.5% (34) | 84.1% (95) | |||
| Other (e.g., Single, Divorced) | 18.7% (14) | 10.5% (4) | 15.9% (18) | |||
| CGAS | 57.27 (6.01) | - | - | - | - | |
| SCARED-P | 31.22 (12.10) | 4.65 (3.80) | 22.70 (16.09) | −16.95 (94.82) | <.001 | |
| SCARED-C | 30.75 (14.07) | 15.19 (9.51) | 25.61 (14.67) | −6.81 (96.60) | <.001 | |
| Co-Sleeping | 8.35 (2) | .02 | ||||
| Usually | 22.7% (17) | 5.3% (2) | 16.8% (19) | |||
| Sometimes | 13.3% (10) | 5.3% (2) | 10.6% (12) | |||
| Rarely | 64% (48) | 89.5% (34) | 72.6% (82) | |||
| WEP | 17.12 (7.71) | 14.24 (8.24) | 16. 11 (7.98) | −1.81 (107) | .07 | |
| WASO (minutes) | 51.14 (26.95) | 44.38 (24.18) | 48.78 (26.11) | −1.29 (107) | .19 | |
| TST (minutes) | 499.92 (39.14) | 504.41 (44.28) | 501.49 (40.86) | .55 (107) | .59 | |
| Sleep Onset | 22:09 (1:04) | 22:02 (0:47) | 22:06 (0:59) | −.62 (107) | .54 | |
| Sleep Offset | 7:19 (0:50) | 7:10 (0:41) | 07:16 (0:47) | −.95 (107) | .35 | |
| TST SD (minutes) | 53.17 (25.98) | 45.15 (17.56) | 50.37 (23.62) | −1.69 (107) | .09 | |
| Sleep Onset SD (minutes) | 39.01 (24.65) | 38.72 (19.39) | 38.91 (22.86) | −.06 (107) | .95 | |
| Sleep Offset SD (minutes) | 42.46 (25.64) | 45.68 (19.13) | 43.58 (23.53) | .68 (107) | .50 | |
Notes. CGAS = Children's Global Assessment Scale. SCARED-P =Screen for Child Anxiety Related Disorders, Parent Report. SCARED-C = Screenfor Child Anxiety Related Disorders, Child Report. WEP = Wake Episodes. WASO= Wake After Sleep Onset. TST = Total Sleep Time. TST SD = StandardDeviation of Total Sleep Time (Intraindividual Variability in SleepDuration). Sleep Onset SD = Standard Deviation of Sleep Onset. Sleep OffsetSD = Standard Deviation of Sleep Offset.
Procedures
Participants were recruited via flyers, local publications, mailings toschool counselors/nurses and pediatrician offices, and at local community eventsfrom both the Washington, D.C. (n = 42) and Houston, TX(n = 71) metropolitan regions. Each child and their parentparticipated in one of two studies conducted by the authors: 1) a research studyon “behavior and emotion” which included a group of childrenwith generalized anxiety disorder (n = 44) and the healthycontrol group (n = 38); and 2) a study offering behavioraltreatment for anxious children (n = 31).1 Advertisem*nts for the formerresearch study targeted both healthy children and children who “worrytoo much.” Advertisem*nts for the latter treatment study targetedfamilies who were “looking for help with their anxious child.”The protocols were approved by an Institutional Review Board at both locations.Parents and children were first consented/assented and were given theopportunity to ask any questions prior to participation. All participantsunderwent an initial assessment that included a set of child and parentquestionnaires and a structured clinical interview. Parents reported ondemographic information, their child’s sleep habits, and theirchild’s anxiety severity. Children also reported on their anxietyseverity. Following this initial appointment, children completed one week ofsleep monitoring using a wrist actigraph (and completion of a sleep diary).During this week, children were instructed to follow their normal sleeproutines. Participants from the research study about “behavior andemotion” were paid $160. Participants from the treatment study wereoffered 16 weeks of free behavioral treatment after the initial assessmentperiod.
Measurement
Clinical Interview
Parents and children participated in separate clinical interviews todetermine diagnostic status using the Anxiety Disorders Interview Schedulefor the DSM-IV for Children and Parents (ADIS-C/P; Silverman and Albano 1996), which is considered thegold standard for diagnosing child anxiety disorders (Lyneham et al. 2007; Silverman et al. 2001) and assesses symptoms across a range ofanxiety, mood, and externalizing disorders. All interviews followedprocedures outlined in the ADIS-C/P Clinician Manual (Albano and Silverman 1996). Children and parents wereinterviewed separately and clinical information and severity ratings werederived from both reporters. As is often the case among anxious youth,children and parents were often discrepant in their severity ratings for aGAD diagnosis (ICC = .02), which were reconciled using establishedprocedural guidelines in the ADIS-C/P manual (pp. 11–14). Clinicianseverity ratings (range = 0–8), which integrate information providedby children and parents, were used to determine severity of each disorder,with ratings of 4 or above signifying clinically-significant impairment.Mean clinician severity ratings for children with GAD ranged from4–8 (M = 6.15, SD = 1.14). Allinterviews were conducted by trained doctoral graduate students andpost-doctoral fellows and were reviewed with a licensed clinicalpsychologist prior to assigning final diagnoses. Interviews werevideo-recorded and 25% were randomly selected for independentratings by a Ph.D. level student in clinical psychology to determinereliability. Interrater reliability for the presence or absence of a GADdiagnosis was excellent for both the research study and the treatment study(kappas = 1.0 and .81, respectively). ADIS-C/P interviews were also used toextract a rating of global functioning for each child using theChildren’s Global Assessment Scale (CGAS), ranging from0–100 (extremely impaired to doing verywell). Some anxious children also met criteria for comorbiddiagnoses, including separation anxiety disorder (n = 13),social anxiety disorder (n = 26), ADHD (n= 7), ODD (n = 2), specific phobia (n =10), panic disorder (n = 1), depressive disorder NOS(n = 3), and tic disorder (n = 1).
Demographics
Parents completed demographic questions regarding their maritalstatus, income, and education, along with their child’srace/ethnicity, age, and gender.
Co-Sleeping
Parents completed the Children’s Sleep Habits Questionnaire(CSHQ; Owens et al. 2000). Parentsreported on how often their child experienced a range of different sleepproblems. Reliability for overall reports of sleep disturbances in thecurrent study (a total score including all sleep problems assessed) was good(α = .84). Co-sleeping was assessed by the question, “Howoften does your child fall asleep in a parent’s or sibling’sbed?” Parents responded by answering on a 3-point scale indicatingwhether their child co-sleeps: 1 (rarely; 0–1 time aweek), 2 (sometimes; 2–4 times aweek), or 3 (usually; 5 or more times a week).
Anxiety Severity
Parents reported on their child’s anxiety severity using theScreen for Child Anxiety Related Disorders: Parent Version (SCARED-P; Birmaher et al. 1999). Children reportedon their anxiety severity using the child version of this measure (SCARED-C)which mirrors the items on the SCARED-P. The SCARED consists of 41 questionsassessing symptoms of panic, GAD, separation anxiety, social phobia, andschool phobia based on DSM-IV classifications. Participants report how trueeach item has been over the past 3 months. Items are measured on a 3-pointLikert scale from 0 (not true or hardly every true) to 2(very true or often true). Ratings were summed tocreate a SCARED-P total score (α = .95) and a SCARED-C total score(α = .93), with higher scores indicating greater anxiety.
Objective Sleep Patterns
Sleep variables were measured using Micro MotionLogger ActigraphSleep Watches (Ambulatory Monitoring, Inc., Ardsley, NY), anaccelerometer-based wrist-worn device that monitors sleep via activity andmovement over the course of one week. Data was collected and stored in oneminute epochs, and was visually inspected to omit epochs where the watch wasnot worn. Data was scored using the zero crossing mode and the Sadehalgorithm (Sadeh et al. 1994), whichis commonly used in pediatric sleep research (Meltzer et al. 2012). Participants were instructed topush an event marking button on the watch to indicate when they got into bedat night and when they got out of bed in the morning, which was used tocalculate the time spent in bed. Families also completed a sleep diary whichprovided a validation check for actigraphy and is considered necessary toaccurately confirm sleep-wake times, identify artifacts, and to clarify anyirregularities in the objective data (Meltzer et al. 2012).
Sleep duration was calculated as the mean number ofminutes spent asleep each night during the actigraphy week. Numberof awakenings was calculated as the average number of blocks ofcontiguous wake epochs each night, and wake after sleeponset was calculated as the average number of minutes spentawake after the onset of sleep each night. Averages were also calculated forthe sleep onset time and sleep offset timeacross the actigraphy week. Finally, variability in sleepduration was calculated as the standard deviation of totalsleep time across the week in minutes for each participant, andvariability in sleep timing was calculated using thestandard deviation of sleep onset times and the standard deviation of sleepoffset times across the week for each participant. The standard deviation iscommonly used for quantifying intraindividual variability in sleep-wakepatterns in children and adolescents (Beckeret al. in press). Three children did not complete the actigraphyweek, and one child was missing 4 nights of data. Based on recommendationsthat a minimum of five nights of actigraphy are necessary to obtain reliablemeasures of sleep in children and adolescents (Acebo et al. 1999), these participants were excludedfrom all analyses using actigraphy.
Analysis Plan
All analyses were conducted using SPSS version 24. Preliminary analyseswere conducted to examine bivariate relationships among variables and to testfor group differences in demographic, sleep, and anxiety variables. To examinerates of co-sleeping among both groups, a chi-square analysis of reportedco-sleeping (i.e., rarely, sometimes, or usually) was conducted, followed by anANCOVA to determine whether group was related to co-sleeping frequency whilecontrolling for covariates. To examine how anxiety severity was related toco-sleeping, we ran a regression model with co-sleeping as the dependentvariable, and parent-reported anxiety symptoms, child-reported anxiety symptoms,and clinician-rated global functioning as independent predictors. To examine the role of co-sleeping in predicting sleep patterns among clinically-anxious youth, we ran a series of regression analyses with co-sleeping as an independent variable, and each sleep outcome as the dependent variable while controlling for covariates.
Results
Preliminary Analyses
Descriptive statistics for each group for are reported in Table 1 and bivariate correlations arereported in Table 2. Anxious childrenexperienced greater parent- and child-reported anxiety, and marginally greaterwake episodes and intraindividual variability in sleep duration. There were noother differences between groups in terms of demographic characteristics orvariables of interest. Older children had lower CGAS scores, and greaterself-reported anxiety. Older children also had a shorter sleep duration and alater sleep onset time. Caucasian children experienced more wake episodes, andearlier and less variable sleep onset and offset times. Parents with higherincome and who were married reported lower rates of co-sleeping and lessintraindividual variability in sleep offset time. Higher income was also relatedto less intraindividual variability in total sleep duration and greater wakeepisodes. Greater parent-reported anxiety was related to later sleep offsettimes and greater variability in total sleep duration, and child-reportedanxiety was related with greater variability in sleep duration.
Table 2
Bivariate Correlations among Variables of Interest for allParticipants
| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1. Age | |||||||||||||||
| 2. Race/Ethnicity | .03 | ||||||||||||||
| 3. Income | .02 | .33*** | |||||||||||||
| 4. Marital Status | −.05 | .21* | .36*** | ||||||||||||
| 5. Co-Sleeping | .06 | −.10 | −.21* | −.25** | |||||||||||
| 6. CGAS | −.28* | .11 | .12 | .10 | −.13 | ||||||||||
| 7. SCARED-P | .05 | −.07 | .02 | −.16 | .47*** | −.43*** | |||||||||
| 8. SCARED-C | .20* | −.03 | −.04 | −.07 | .28** | −.36** | .41*** | ||||||||
| 9. WEP | −.08 | .23* | .28* | .09 | −.09 | .05 | .10 | −.06 | |||||||
| 10. WASO | −.11 | .15 | .16 | .10 | −.07 | .08 | .08 | −.13 | .86** | ||||||
| 11. TST | −.34*** | .11 | .05 | −.05 | −.09 | .06 | −.06 | −.05 | −.06 | −.18 | |||||
| 12. Sleep Onset | .31** | −.19* | −.15 | −.01 | .33*** | −.17 | .17 | .16 | −.29** | −.31** | −.48*** | ||||
| 13. Sleep Offset | .02 | −.05 | −.07 | .01 | .30** | −.13 | .21* | .07 | .08 | .02 | .15 | .68*** | |||
| 14. TST SD | −.04 | −.11 | −.21* | −.14 | .35*** | −.08 | .28* | .26** | −.05 | −.09 | −.05 | .22* | .18 | ||
| 15. Sleep Onset SD | .17 | −.25** | −.18 | −.12 | .09 | −.16 | −.01 | .10 | −.18 | −.11 | −.17 | .37*** | −.08 | −.08 | |
| 16. Sleep Offset SD | .10 | −.28** | −.27** | −.24* | .23* | −.04 | .04 | .09 | −.21* | −.19 | −.19* | .53*** | .14 | .14 | .49*** |
Notes.
***p < .001,
**p < .01,
*p < .05.
CGAS = Children’s Global Assessment Scale (reported for theanxious group only). SCARED-P = Screen for Child Anxiety Related Disorders,Parent Report. SCARED-C = Screen for Child Anxiety Related Disorders, ChildReport. WEP = Wake Episodes. WASO = Wake After Sleep Onset. TST = TotalSleep Time. TST SD = Standard Deviation of Total Sleep Time (IntraindividualVariability in Sleep Duration). Sleep Onset SD = Standard Deviation of SleepOnset. Sleep Offset SD = Standard Deviation of Sleep Offset. Co-Sleeping iscoded as 1 = usually, 0 = rarely or sometimes. Race/ethnicity is dummy codedas 1 = white/Caucasian, 0 = other. Marital status is dummy coded as 1 =married, 0 = other. All CGAS correlations are presented for the anxiousgroup only.
Aim 1: Co-Sleeping in Clinically-Anxious and Healthy Children
A chi-square analysis indicated that anxious children co-slept moreoften than healthy controls (Cramer’s V = .27). Parents of anxiouschildren were significantly more likely to report that their child usually orsometimes co-slept compared to healthy controls. These results are displayed inTable 1 and Figure 1. After controlling for age, income, childrace/ethnicity, and parent marital status, findings were similar, with anxiouschildren co-sleeping more often than healthy controls, F(1,105) = 9.26, p = .003, 95% CI [−.76,−.16], partial η2 = .08.
To examine whether severity of anxiety pathology (assessed viaparent-reported anxiety symptoms, child-reported anxiety symptoms, andclinician-rated global functioning ratings) predicted co-sleeping in the anxiousgroup specifically, we ran a series of hierarchical regression analyses usingage, income, child race/ethnicity, and parent marital status as covariates.Greater frequency of co-sleeping was significantly predicted by parent-reportedanxiety severity and by child-reported anxiety severity, but not byclinician-rated global functioning. See Table3.
Table 3
Anxiety Severity Predicting Co-Sleeping in Anxious Children
| Co-Sleeping | |||||
|---|---|---|---|---|---|
| Adj. R2 | β | b | SE | 95% CI | |
| .14* | |||||
| Age | −.02 | −.01 | .06 | −.13, .11 | |
| Race/Ethnicity | .01 | .01 | .21 | −.40, .42 | |
| Income | −.17 | −10 | .07 | −.24, .05 | |
| Parent Marital Status | −.08 | −.18 | .26 | −.70, .35 | |
| SCARED-P | .37** | .03 | .01 | .01, .04 | |
| SCARED-C | .21+ | .01 | .01 | −.002, .03 | |
| CGAS | .09 | .01 | .02 | −.03, .06 | |
Notes.
**p < .01,
*p < .05,
+p < .10.
Standardized and unstandardized and standardized estimates arereported.
CGAS = Children’s Global Assessment Scale. SCARED-P = Screenfor Child Anxiety Related Disorders, Parent Report. SCARED-C = Screen forChild Anxiety Related Disorders, Child Report. Race/ethnicity is dummy codedas 1 = white/Caucasian, 0 = other. Marital status is dummy coded as 1 =married, 0 = other.
Aim 2: Co-Sleeping and Objective Sleep Patterns in Anxious Youth
To examine relationships between co-sleeping and objective sleepparameters, a series of hierarchical regression models were conducted withco-sleeping as the independent variable (and age, income, child race/ethnicity,and parent marital status as covariates). Only clinically-anxious children wereincluded in the models given the low overall incidence of co-sleeping behavioramong controls. Clock time variables (i.e., sleep onset and offset times) wereconverted to minutes and centered around the sample mean to aid in theinterpretability of regression coefficients. When examining mean levels of sleepdisturbances during the actigraphy week, co-sleeping was not related to thenumber of awakenings, wake after sleep onset, or to total sleep time (see Table 3). However, greater co-sleeping wasrelated to mean levels of sleep timing, including later sleep onset times andlater sleep offset times (see Table 4).When examining intraindividual variability across the week, co-sleeping wasassociated with greater variability in the total sleep time, but not tovariability in sleep timing. These results are displayed in Table 5.
Table 4
Co-Sleeping Predicting Average Objective Sleep Parameters in AnxiousChildren
| WEP | WASO | TST | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| β | b | SE | 95% CI | β | b | SE | 95% CI | β | b | SE | 95% CI | |
| Age | −.19 | −.89 | .53 | −1.95, .18 | −.24* | −3.94 | 1.89 | −7.70, −.17 | −.36** | −8.52 | 2.66 | −13.82, −3.31 |
| Race/Ethnicity | .06 | .86 | 1.83 | −2.81, 4.53 | .01 | .67 | 6.45 | −12.29, 13.64 | .20 | 15.46 | 9.14 | −2.80, 33.72 |
| Income | .26 | 1.36 | .74 | −.12, 2.83 | .09 | 1.57 | 2.61 | −3.64, 6.78 | .09 | 2.38 | 3.68 | −4.97, 9.72 |
| Parent Marital Status | .11 | 2.17 | 2.54 | −2.90, 7.25 | .17 | 11.71 | 8.97 | −6.20, 29.62 | −.30* | −30.58 | 12.63 | −55.81, −5.35 |
| Co-Sleeping | −.08 | −.73 | 1.13 | −2.99, 1.54 | −.15 | −4.70 | 4.00 | −12.70, 3.29 | −.01 | −.32 | 5.67 | −11.58, 10.93 |
| Adj. R2 | .12* | .10* | .16** | |||||||||
Notes.
**p < .01,
*p < .05.
Standardized and unstandardized estimates are reported.
WEP = Wake Episodes. WASO = Wake After Sleep Onset. TST = TotalSleep Time. Race/ethnicity is dummy coded as 1 = white/Caucasian, 0 = other.Marital status is dummy coded as 1 = married, 0 = other.
Table 5
Co-Sleeping Predicting Average Objective Sleep Timing in AnxiousChildren
| Sleep Onset Time | Sleep Offset Time | |||||||
|---|---|---|---|---|---|---|---|---|
| β | b | SE | 95% CI | β | b | SE | 95% CI | |
| Age | .27* | 10.64 | 4.14 | 2.37, 19.90 | −.08 | −2.34 | 3.53 | −9.38, 4.70 |
| Race/Ethnicity | −.17 | −21.35 | 14.24 | −49.80, 7.10 | −.05 | −5.33 | 12.13 | −29.56, 18.91 |
| Income | −.28* | −12.34 | 5.73 | −23.78, −.91 | −.24 | −8.37 | 4.88 | −18.11, 1.37 |
| Parent Marital Status | .25* | 41.21 | 19.68 | 1.90, 80.52 | .19 | 24.44 | 16.76 | −9.04, 57.92 |
| Co-Sleeping | .28* | 21.64 | 8.78 | 4.10, 39.18 | .30* | 17.89 | 7.48 | 2.95, 32.83 |
| Adj. R2 | .25*** | .12* | ||||||
Notes.
***p < .001,
*p < .05.
Standardized and unstandardized estimates are reported.
Race/ethnicity is dummy coded as 1 = white/Caucasian, 0 = other.Marital status is dummy coded as 1 = married, 0 = other.
Discussion
The current study fills gaps in the limited body of research examining thepresence and associated sleep patterns of co-sleeping in anxious school-agedchildren. Building on research showing anxious youth often struggle with sleepproblems, we investigated the frequency of co-sleeping in both healthy andclinically anxious school-aged children and relationships between co-sleeping andanxiety severity using multiple reporters. We also examined sleep-related correlatesof co-sleeping in this anxious sample using objective (i.e., actigraphy-based) sleepassessments. As expected, we found anxious children to co-sleep significantly morethan healthy children, with more than 1 in 3 anxious children co-sleeping usually orsometimes. Importantly, co-sleeping was even more common among youth with greateranxiety severity. The high rates reported in this sample are remarkable given therelative lack of co-sleeping research in this population. Importantly, co-sleepingwas also related to more problematic sleep patterns in anxious youth (i.e., greaterintraindividual variability in sleep duration and shifted bed and wake times).Overall, this study provides novel information regarding the occurrence ofco-sleeping in school-aged anxious youth, and suggests that frequent co-sleeping maybe one factor associated with sleep problems in this population. Problematicsleep-onset associations may be a malleable treatment target for school-agedchildren (Paine and Gradisar 2011), and thusthese findings might be used to inform intervention and prevention efforts.
Our finding that over one third of anxious children sleep in a parent orsibling’s bed (at least a couple of times a week) is in line with high ratesof bedtime resistance and refusal to sleep alone found in other studies (Alfano et al. 2007). Further, among anxiousyouth, co-sleeping was linked with more severe forms of anxiety (based on bothparent and child reports), fitting the view that co-sleeping is reactionary inresponse to heightened fear and arousal surrounding sleep (Alfano et al. 2010; Forbes etal. 2006; Hansen et al. 2011).Although we did not directly assess family preferences for sleeping arrangements,overall rates of sleep complaints among anxious youth and their parents (Alfano et al. 2007; Alfano et al. 2010) along with the frequency with whichco-sleeping is identified by parents as a specific goal of treatment (Paine and Gradisar 2011), suggests thatco-sleeping is often reactive in this population. Generally, co-sleeping declines aschildren enter the school-aged years (BaHammam et al.2008; Cortesi et al. 2004; Jenni et al. 2005; Li et al. 2009). Inschool-aged children with anxiety disorders however, persistent nighttime fears andproblems initiating sleep may give rise to parental accommodation in the form ofco-sleeping (Thompson-Hollands et al.2014).
Precise mechanisms of sleep disturbances among anxious individuals arelargely unknown; however, our study identified co-sleeping as a potentiallyimportant yet unexplored correlate of sleep problems among clinically-anxious youth.It is important to note, however, that we did not find significant differences inactigraphy-measured sleep indicators between healthy and anxious children (with theexception of a marginal difference in wake episodes and variability in total sleepduration). This is in line with many studies investigating sleep in anxious youth,which often find high rates of subjectively reported sleep problems, but fail tofind differences compared to healthy youth using objective methodology (McMakin and Alfano 2016). However, our findingssuggest that objectively measured sleep disruptions may be greatest among the subsetof anxious youth who frequently co-sleep. These findings are not surprising givenprior research linking co-sleeping in healthy school-aged youth to a range of sleepdifficulties.
In particular, we found that anxious youth who co-slept more often were alsomore likely to experience a greater shift in their sleep timing (i.e., later sleeponset and offset times). Children with anxiety often resist going to bed (Alfano et al. 2010), which can create stress forboth parents and children, prolong bedtime routines (Meltzer 2010), and culminate in parents allowing children to co-sleep.Paradoxically, children who avoid nighttime anxiety by co-sleeping may disrupt theirsleep even more via shifting their bedtime later in order to coincide withparents’ bedtime. A shift in sleep timing during the school-aged years maycontribute to and potentially exacerbate the reliable, biologically-based circadiandelay that occurs with the onset of puberty (Burgessand Eastman 2004; Carskadon et al.1993). As a later circadian phase in adolescence is associated withgreater emotional distress and depressive symptoms (McGlinchey and Harvey 2015; Short et al.2013), identifying these problematic patterns early, when they may bemore malleable (Takeuchi et al. 2001), seemscritical.
In addition, although we did not find relationships with overall sleepduration, we did find that co-sleeping was related to greater intraindividualvariability in sleep duration among anxious children. Few studies have consideredthe impact of co-sleeping on the consistency of children’s sleep. It ispossible that the later bedtimes that we observed among our co-sleeping anxiousyouth resulted in less sleep throughout the week, necessitating longer“catch up” sleep periods on weekends and greater variability insleep duration from weekdays to weekends. Across ages, less consistency ofsleep-wake patterns corresponds with more behavioral, academic, and health-relatedproblems (Becker et al. in press; Thoman 1990; Wolfson and Carskadon 1998), which can in turn maintain anxiety symptomsover time. Children with attention deficit/hyperactivity disorder (ADHD), forexample, exhibit more variable sleep-wake patterns than healthy children (Gruber et al. 2000). As both ADHD and anxietycan be viewed as disorders of arousal regulation (Brown and McMullen 2001), greater night-to-night sleep variability couldbe an antecedent as much as a consequence of these conditions. Indeed, sleeprhythmicity (i.e., regularity of sleep schedules) in early childhood has been shownto precede the onset of anxiety disorders in adolescence (Ong et al. 2006). Thus, a focus on sleep variability, inaddition to total sleep duration, should be a focus of behavioral sleepinterventions for children, particularly those with elevated levels of anxiety.
Limitations and Future Directions
Despite their important clinical implications, our findings are notwithout limitations. First, although we used an objective measure of sleep andmultiple reporters of anxiety severity, our co-sleeping assessment only capturedbed sharing with a parent or sibling. Further, we only assessed parentalretrospective report of co-sleeping, which may not have accurately reflected howoften children co-slept during the actigraphy week. Future research shouldtherefore more comprehensively assess co-sleeping. Importantly, a subset of ouranxious participants were treatment-seeking, and prior research suggests thatanxious youth receiving treatment are more impaired than anxious youths whoparticipate in non-treatment-related research (De Los Reyes et al. 2017). Thus, findings require replication in aclinic-based sample of anxious youth, especially considering some of our mainvariables of interest (e.g., reports of anxiety severity, sleep variability)differed between the two samples.1 Varying compensation across our two samples should also benoted (i.e., research-seeking families were provided with monetary compensationwhile treatment-seeking families received free treatment). Finally, although ourmodels examined anxiety as a predictor of co-sleeping and co-sleeping as apredictor of sleep-wake patterns, the cross-sectional nature of these datapreclude our ability to determine directionality. Given bidirectionalrelationships between sleep problems and anxiety symptoms (McMakin and Alfano 2016), as well as co-sleeping andsleep-related difficulties (Mileva-Seitz et al.2017), it is possible that the sleep problems in our anxious youthmay have preceded their co-sleeping. Longitudinal research can provide moreinformation on the directional nature of these relationships.
Conclusions
In sum, the current research provides evidence that co-sleeping iscommon in clinically-anxious school-aged youth and corresponds with more severeforms of anxiety. Building on research revealing a majority of children withanxiety disorders experience sleep-related problems, these findings suggest thatfrequent co-sleeping in this population may be one factor that disrupts healthysleep patterns. At the same time, we urge clinicians as well as researchers tobroaden conceptualizations of problematic sleep to include the timing of whensleep occurs and the night-to-night inconsistency of sleep schedules. The use ofa one-time report of a child’s bed/wake times and/or total sleepduration ignores the wealth of evidence linking inconsistent sleep patterns withdeleterious child outcomes. Moreover, based on the reciprocal relationshipshared by anxiety and sleep, frequent co-sleeping might serve as a risk and/ormaintaining factor of anxiety pathology vis-à-vis disrupted sleeppractices. In such cases, co-sleeping should be considered a target forintervention.
Table 6
Co-Sleeping Predicting Objective Intraindividual Variability in SleepDuration and Timing in Anxious Children
| TST SD | Sleep Onset SD | Sleep Offset SD | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| β | b | SE | 95% CI | β | b | SE | 95% CI | β | b | SE | 95% CI | |
| Age | −.04 | −.55 | 1.70 | −3.94, 2.84 | .22 | 3.31 | 1.79 | −.28, 6.89 | .03 | .51 | 1.78 | −3.04, 4.07 |
| Race/Ethnicity | −.04 | −2.11 | 6.04 | −14.18, 9.96 | −.19 | −9.15 | 6.17 | −21.48, 3.17 | −.14 | −7.18 | 6.12 | −19.42, 5.05 |
| Income | −.35** | −6.05 | 2.28 | −10.61, −1.48 | −.13 | −2.24 | 2.48 | 7.20, 2.72 | −.16 | −2.67 | 2.46 | −7.59, 2.25 |
| Parent Marital Status | .07 | 4.76 | 8.19 | −11.59, 21.12 | .02 | 1.15 | 8.52 | −15.88, 18.19 | −.14 | −9.19 | 8.46 | −26.10, 7.72 |
| Co-Sleeping | .29* | 8.91 | 3.56 | 1.76, 16.05 | −.04 | 1.22 | 3.81 | −8.82, 6.38 | .14 | 4.31 | 3.78 | −3.23, 11.86 |
| Adj. R2 | .19** | .03 | .09* | |||||||||
Notes.
**p < .01,
*p < .05.
TST SD = Standard Deviation of Total Sleep Time (IntraindividualVariability in Sleep Duration). Sleep Onset SD = Standard Deviation of SleepOnset. Sleep Offset SD = Standard Deviation of Sleep Offset. Race/ethnicityis dummy coded as 1 = white/Caucasian, 0 = other. Marital status is dummycoded as 1 = married, 0 = other.
Acknowledgments
This project was funded by grant K23MH081188 from the National Institute ofMental Health (PI: C. Alfano), University of Houston Small Grants Program (Co-PIs:C. Alfano & M. Clementi), and Texas Psychological Foundation GraduateResearch Proposal Grant (PI: M. Clementi).
Footnotes
1There were several significant differences between the research-seekingand treatment-seeking samples on variables of interest. The research-seekingsample had greater CGAS scores compared to the treatment-seeking sample[t(73) = 3.34, p = .001;Mresearch = 59.09,SDresearch = 5.93,Mtreatment = 54.68,SDtreatment = 5.18]. The treatment-seekingsample had greater parent-reported SCARED scores [t(70) =−2.29, p = .025; Mresearch= 28.46, SDresearch = 12.37,Mtreatment = 34.87,SDtreatment = 10.87], child-reported SCAREDscores [t(71) = −2.45, p = .017;Mresearch = 27.40,SDresearch = 13.37,Mtreatment = 35.29,SDtreatment = 13.91], and variability intotal sleep duration [t(33.37) = −4.36,p < .001; Mresearch =42.75, SDresearch = 16.07,Mtreatment = 70.85,SDtreatment = 30.48]. Participants in thetreatment-seeking sample also had a later average sleep onset time[t(34.52) = −2.13, p = .041;Mresearch = 21:55,SDresearch = 0:43,Mtreatment = 22:32,SDtreatment = 1:25]. Importantly, thesignificance and direction of all findings generally remained the same whenaccounting for the study as a covariate, with the exception of the relationshipbetween co-sleeping and variability in total sleep duration, which dropped tomarginal in the regression model (p = .06). Although the smallsample of anxious children in each study limit our ability to draw firmconclusions based on these findings, these results generally suggest that thetreatment-seeking sample presented with more severe symptoms than theresearch-seeking sample.
References
- Acebo C, Sadeh A, Seifer R, Tzischinsky O, Wolfson AR, Hafer A, Carskadon M. Estimating sleep patterns with activity monitoring in childrenand adolescents: How many nights are necessary for reliablemeasures? Sleep. 1999;22:95–103. [PubMed] [Google Scholar]
- Albano AM, Silverman WK. The Anxiety Disorders Interview Schedule for Children for DSM-IV:Clinician Manual (Child and Parent Versions) Psychological Corporation; San Antonio, TX: 1996. [Google Scholar]
- Alfano CA, Ginsburg GS, Kingery JN. Sleep-related problems among children and adolescents withanxiety disorders. Journal of the American Academy of Child and AdolescentPsychiatry. 2007;46:224–232. [PubMed] [Google Scholar]
- Alfano CA, Patriquin MA, De Los Reyes A. Subjective – objective sleep comparisons anddiscrepancies among clinically-anxious and healthy children. Journal of Abnormal Child Psychology. 2015;43:1343–1353. [PMC free article] [PubMed] [Google Scholar]
- Alfano CA, Pina AA, Zerr AA, Villalta IK. Pre-sleep arousal and sleep problems of anxiety-disorderedyouth. Child Psychiatry and Human Development. 2010;41:156–167. [PMC free article] [PubMed] [Google Scholar]
- BaHammam A, Alameri H, Hersi A. Cosleeping and its correlates in Saudi school-agedchildren. International Journal of Behavioral Medicine. 2008;15:336–340. [PubMed] [Google Scholar]
- Birmaher B, Brent DA, Chiappetta L, Bridge J, Monga S, Baugher M. Psychometric properties of the Screen for Child Anxiety RelatedEmotional Disorders (SCARED): A replication study. Journal of the American Academy of Child and AdolescentPsychiatry. 1999;38:1230–1236. [PubMed] [Google Scholar]
- Becker SP, Sidol CA, Van Dyk TR, Epstein JN, Beebe DW. Intraindividual variability of sleep/wake patterns in relation tochild and adolescent functioning: A systematic review. Sleep Medicine Reviews. (in press). Online ahead of print. [PMC free article] [PubMed] [Google Scholar]
- Beidel DC, Alfano CA. Childhood Anxiety Disorders: A Guide to Research and Treatment. 2. New York, NY: Taylor & Francis/Routledge MentalHealth; 2011. [Google Scholar]
- Blader JC, Koplewicz HS, Abikoff H, Foley C. Sleep problems of elementary school children. A communitysurvey. Archives of Pediatrics and Adolescent Medicine. 1997;151:473–480. [PubMed] [Google Scholar]
- Brown TE, McMullen WJ., Jr Attention deficit disorders and sleep/arousaldisturbance. Annals of the New York Academy of Sciences. 2001;931:271–286. [PubMed] [Google Scholar]
- Buckhalt JA, El-Sheikh M, Keller P. Children’s sleep and cognitive functioning: race andsocioeconomic status as moderators of effects. Child Development. 2007;78:213–231. [PubMed] [Google Scholar]
- Burgess HJ, Eastman CI. Early versus late betimes phase shift the human dim lightmelatonin rhythm despite a fixed morning lights on time. Neuroscience Letters. 2004;356:115–118. [PMC free article] [PubMed] [Google Scholar]
- Carskadon MA, Viera C, Acebo C. Association between puberty and delayed phasepreference. Sleep. 1993;16:258–262. [PubMed] [Google Scholar]
- Carthy T, Horesh N, Apter A, Edge MD, Gross JJ. Emotional reactivity and cognitive regulation in anxiouschildren. Behaviour Research and Therapy. 2010;48:384–393. [PubMed] [Google Scholar]
- Cortesi F, Giannotti F, Sebastiani T, Vagnoni C. Cosleeping and sleep behavior in Italian school-agedchildren. Journal of Developmental and Behavioral Pediatrics. 2004;25:28–33. [PubMed] [Google Scholar]
- Cortesi F, Giannotti F, Sebastiani T, Vagnoni C, Marioni P. Cosleeping versus solitary sleeping in children with bedtimeproblems: Child emotional problems and parental distress. Behavioral Sleep Medicine. 2008;2:89–105. [PubMed] [Google Scholar]
- Cowie J, Alfano CA, Patriquin MA, Reynolds KC, Talavera D, Clementi MA. Addressing sleep in children with anxietydisorders. Sleep Medicine Clinics. 2014;9:137–148. [Google Scholar]
- De Los Reyes A, Alfano CA, Clementi MA, Viana A. Are the clinical characteristics of anxious youths participatingin non-treatment-related research comparable to those of youths receivingtreatment? Child Youth Car Forum. 2017;46:1–17. [Google Scholar]
- Doane LD, Gress-Smith JL, Breitenstein RS. Multi-method assessments of sleep over the transition to collegeand the associations with depression and anxiety symptoms. Journal of Youth and Adolescence. 2015;44:389–404. [PubMed] [Google Scholar]
- Emens J, Lewy A, Kinzie JM, Arntz D, Rough J. Circadian misalignment in major depressivedisorder. Psychiatry Research. 2009;168:259–261. [PubMed] [Google Scholar]
- Fletcher FE, Conduit R, Foster-Owens MD, Rinehart NJ, Rajaratnam SMW, Cornish KM. The association between anxiety symptoms and sleep in school-agedchildren: a combined insight from the Children’s Sleep HabitsQuestionnaire and actigraphy. Behavioral Sleep Medicine. (in press). Online ahead of print. [PubMed] [Google Scholar]
- Forbes EE, Williamson DE, Ryan ND, Birmaher B, Axelson DA, Dahl RE. Peri-sleep-onset cortisol levels in children and adolescents withaffective disorders. Biological Psychiatry. 2006;59:24–30. [PMC free article] [PubMed] [Google Scholar]
- Fuligni AJ, Hardway C. Daily variation in adolescents’ sleep, activities, andpsychological well-being. Journal of Research on Adolescence. 2006;16:353–378. [Google Scholar]
- Gruber R, Sadeh A, Rviv A. Instability of sleep patterns in children withattention-deficit/hyperactivity disorder. Journal of the American Academy of Child and AdolescentPsychiatry. 2000;38:495–501. [PubMed] [Google Scholar]
- Goldberg WA, Keller MA. Parent-infant co-sleeping: why the interest andconcern? Infant and Child Development. 2007;16:331–339. [Google Scholar]
- Hansen BH, Skirbekk B, Oerbeck B, Richter J, Kristensen H. Comparison of sleep problems in children with anxiety andattention deficit/hyperactivity disorders. European Child and Adolescent Psychiatry. 2011;20:321–330. [PubMed] [Google Scholar]
- Hayes MJ, Parker KG, Sallinen B, Davare AA. Bed sharing, temperament, and sleep disturbance in earlychildhood. Sleep. 2001;24:657–662. [PubMed] [Google Scholar]
- Hudson JL, Gradisar M, Gamble A, Schniering CA, Rebelo I. The sleep patterns and problems of clinically anxiouschildren. Behaviour Research and Therapy. 2009;47:339–344. [PubMed] [Google Scholar]
- Jain S, Romack R, Jain R. Bed sharing in school-age children-clinical and socialimplications. Journal of Child and Adolescent Psychiatric Nursing. 2011;24:185–189. [PubMed] [Google Scholar]
- Jenni OG, Fuhrer HZ, Iglowstein I, Molinari L, Largo RH. A longitudinal study of bed sharing and sleep problems amongSwiss children in the first 10 years of life. Pediatrics. 2005;115:233–240. [PubMed] [Google Scholar]
- Jiang Y, Chen W, Spruyt K, Sun W, Wang Y, Li S, Shen X, Wang G, Jiang F. Bed-sharing and related factors in earlyadolescents. Sleep Medicine. 2016;17:75–80. [PubMed] [Google Scholar]
- Keller MA, Goldberg WA. Co-sleeping: Help or hindrance for young children’sindependence? Infant and Child Development. 2004;13:369–388. [Google Scholar]
- Latz S, Wolf AW, Lozoff B. Cosleeping in context: Sleep practices and problems in youngchildren in Japan and the United States. Archives of Pediatrics and Adolescent Medicine. 1999;153:339–346. [PubMed] [Google Scholar]
- Liu X, Liu L, Wang R. Bed sharing, sleep habits, and sleep problems among Chineseschool-aged children. Pediatrics. 2003;26:839–844. [PubMed] [Google Scholar]
- Lozoff B, Wolf AW, Davis NS. Cosleeping in urban families with young children in the UnitedStates. Pediatrics. 1984;74:171–182. [PubMed] [Google Scholar]
- Lyneham HJ, Abbott MJ, Rapee RM. Interrater reliability of the Anxiety Disorders InterviewSchedule for DSM-IV: Child and parent versions. Journal of the American Academy of Child and AdolescentPsychiatry. 2007;46:731–736. [PubMed] [Google Scholar]
- McGlinchey EL, Harvey AG. Risk behaviors and negative health outcomes for adolescents withlate bedtimes. Journal of Youth and Adolescence. 2015;44:478–488. [PMC free article] [PubMed] [Google Scholar]
- McKenna JJ, Volpe LE. Sleeping with baby: An internet-based sampling of parentalexperiences, choices, perceptions, and interpretations in a Westernindustrialized context. Infant and Child Development. 2007;16:359–385. [Google Scholar]
- McMakin DL, Alfano CA. Sleep and anxiety in late childhood and earlyadolescence. Current Opinion in Psychaitry. 2016;28:483–489. [PMC free article] [PubMed] [Google Scholar]
- Meltzer LJ. Clinical management of behavioral insomnia of childhood:Treatment of bedtime problems and night wakings in youngchildren. Behavioral Sleep Medicine. 2010;8:172–189. [PubMed] [Google Scholar]
- Meltzer LJ, Montgomery-Downs HE, Insana SP, Walsh CM. Use of actigraphy for assessment in pediatric sleepresearch. Sleep Medicine Reviews. 2012;16:463–475. [PMC free article] [PubMed] [Google Scholar]
- Mileva-Setiz VR, Bakermans-Kranenburg MJ, Battaini C, Luijk MPCM. Parent-child bed-sharing: The good, the bad, and the burden ofevidence. Sleep Medicine Reviews. 2017;32:4–27. [PubMed] [Google Scholar]
- Mindell JA, Meltzer LJ, Carskadon MA, Chervin RD. Developmental aspects of sleep hygiene: Findings from the 2004National Sleep Foundation Sleep in America Poll. Sleep Medicine. 2009;10:771–779. [PubMed] [Google Scholar]
- Ong SH, Wickramaratne P, Tang M, Weissman MM. Early childhood sleep and eating problems as predictors ofadolescent and adult mood and anxiety disorders. Journal of Affective Disorders. 2006;96:1–8. [PubMed] [Google Scholar]
- Owens JA. Sleep in children: Cross-cultural perspectives. Sleep and Biological Rhythms. 2004;2:165–173. [Google Scholar]
- Owens JA, Spirito A, McGuinn M. The Children’s Sleep Habits Questionnaire (CSHQ):Psychometric properties of a survey instrument for school-agedchildren. Sleep. 2000;23:1–9. [PubMed] [Google Scholar]
- Paine S, Gradisar M. A randomised controlled trial of cognitive-behaviour therapy forbehavioural insomnia of childhood in school-aged children. Behavior Research and Therapy. 2011;49:379–388. [PubMed] [Google Scholar]
- Peterman JS, Carper MM, Kendall PC. Anxiety disorders and comorbid sleep problems in school-agedyouth: A review and future research directions. Child Psychiatry and Human Development. 2015;46:376–392. [PubMed] [Google Scholar]
- Rath FH, Jr, Okum ME. Parents and children sleeping together: Co-sleeping prevalenceand concerns. American Journal of Orthopsychiatry. 1995;65:411–418. [PubMed] [Google Scholar]
- Sadeh A, Anders TF. Infant sleep problems: Origins, assessment,interventions. Infant Mental Health Journal. 1993;14:17–34. [Google Scholar]
- Sadeh A, Sharkey KM, Carskadon MA. Activity-based sleep—wake identification: an empiricaltest of methodological issues. Sleep. 1994;17:201–207. [PubMed] [Google Scholar]
- Silverman WK, Albano A. The Anxiety Disorders Interview Schedule for Children-IV (Child andparent versions) San Antonio: Psychological Corporation; 1996. [Google Scholar]
- Silverman WK, Saaverdra LM, Pina AA. Test-retest reliability of anxiety symptoms and diagnoses withthe Anxiety Disorders Interview Schedule for DSM-IV: Child and parentversons. Adolescent Psychiatry. 2001;40:937–944. [PubMed] [Google Scholar]
- Short MA, Gradisar M, Lack LC, Wright HR. The impact of sleep on adolescent depressed mood, alertness andacademic performance. Journal of Adolescence. 2013;36:1025–1033. [PubMed] [Google Scholar]
- Smedje H, Broman JE, Hetta J. Short-term prospective study of sleep disturbances in5–8-year-old children. Acta Paediatr. 2001;90:1456–1463. [PubMed] [Google Scholar]
- Storch EA, Geffken GR, Merlo LJ, Jacob ML, Murphy TK, Goodman WK, Larson MJ, Fernandez M, Grabill K. Family accommodation in pediatric obsessive-compulsivedisorder. Journal of Clinical Child and Adolescent Psycholology. 2007;36:207–216. [PubMed] [Google Scholar]
- Suveg C, Zeman J. Emotion regulation in children with anxietydisorders. Journal of Clinical Child and Adolescent Psychology. 2004;33:750–759. [PubMed] [Google Scholar]
- Takeuchi H, Inoue M, Watanabe N, Yamash*ta Y, Hamada M, Kadota G, Harada T. Parental enforcement of bedtime during childhood modulatespreference of Japanese junior high school students for eveningnesschronotype. Chronobiology International. 2001;18:823–829. [PubMed] [Google Scholar]
- Thelen E. Self-organization in developmental processes: Can systemsapproaches work? In: Johnson MH, editor. Brain development and cognition. Oxford, England: Blackwell; 1993. pp. 555–592. [Google Scholar]
- Thoman EB. Sleeping and waking states in infants: A functionalperspective. Neuroscience and Biobehavioral Reviews. 1990;14:93–107. [PubMed] [Google Scholar]
- Thoman EB. Co-sleeping, an ancient practice: Issues of the past and present,and possibilities for the future. Sleep Medicine Reviews. 2006;10:407–417. [PubMed] [Google Scholar]
- Thompson-Hollands J, Kerns CE, Pincus DB, Comer JS. Parental accomodation of child anxiety and related symptoms:Range, impact, and correlates. Journal of Anxiety Disorders. 2014;28:765–773. [PMC free article] [PubMed] [Google Scholar]
- Wolfson AR, Carskadon MA. Sleep schedules and daytime functioning inadolescents. Child Development. 1998;69:875–887. [PubMed] [Google Scholar]
