Does Disturbed Sleeping Precede Symptoms of Anxiety or Depression in Toddlers?The Generation R Study

P AULINE W.J ANSEN,P H D,N ATHALIE S.S ARIDJAN,MD,A LBERT H OFMAN,MD,P H D,V INCENT W.V.J ADDOE,MD,P H D, F RANK C.V ERHULST,MD,P H D,AND H ENNING T IEMEIER,MD,P H D

Objective:To examine whether sleep problems in infancy and early toddlerhood precede symptoms of anxiety or depression at3 years.Methods:Data on specific sleep problems at2months and24months were available for4,782children participating in a population-based cohort in The Netherlands.The Child Behavior Checklist for toddlers containing the Anxious/Depressed syndrome scale was assessed at36months.We adjusted the logistic regression analyses for several confounding factors;the analyses with sleep problems at24months were additionally adjusted for preexisting anxiety or depressive symptoms(at18 months).Results:Dyssomnia and parental presence during sleep onset at2months and24months were associated with anxiety

or depressive symptoms at3years(e.g.,parental presence:odds ratio

2months ,1.22;95%confidence interval,1.04–1.44;odds ratio

24

months,1.58;95%confidence interval,1.30–1.92).Parasomnia,short sleep duration,and absence of set bedtime at24months,but not at2months,also preceded anxiety or depressive symptoms.These significant associations were not due to children’s anxiety or depressive symptoms at18months.Rhythmicity and co-sleeping were not associated with later anxiety or depressive symptoms. Additional analyses provided little evidence for a bidirectional association with anxiety or depressive symptoms preceding later sleep problems.Conclusions:Our findings highlight the importance of sleep problems early in life,because different sleep problems are associated with the frequency of anxiety or depressive symptoms.Therefore,healthcare practitioners must be particularly attentive to these problems in young children.Future research should address possible mechanisms underlying the association between disturbed sleeping and anxiety or depressive symptoms.Key words:sleep problems,anxiety,depression.

CBCL/11?2-5?Child Behavior Checklist for toddlers;OR?odds ratio;95%CI?95%confidence interval.

INTRODUCTION

A nxiety disorders and depression are among the most

common psychiatric diagnoses in children and adoles-cents with prevalence rates estimated between2%and10% (1–3).It is well established that children with these disor-ders often have comorbid sleep problems.Several studies in clinical samples indicated that anxious and depressed chil-dren are troubled sleepers who are characterized by diffi-culties in falling asleep,reluctance to sleep alone,frequent awakenings,nightmares,and overtiredness(4–7).Alfano and colleagues(5)indicated that88%of children with an anxiety disorder experienced at least one sleep-related problem.Sleep problems are a common and important symptom of anxiety disorders and depression.

Yet,there is emerging evidence from longitudinal research in the general population that disturbed sleep might also predispose children to anxiety and depression:It was repeat-edly indicated that sleep problems during childhood preceded later anxiety disorders and depression with follow-up periods varying from1year to15years(8–14).Several studies (10,12,13),however,did not account for preexisting anxiety or depressive symptoms in children.Thus,in these studies,it cannot be ruled out that symptoms of anxiety or depression caused sleep problems before the baseline measure was ob-tained.In addition,previous research was mainly conducted in school-aged children.This further complicates our ability to understand the interrelationship between sleep and anxiety or depression,as their development often begins early in life(2). Another feature of most longitudinal research on sleep prob-lems and anxiety or depressive symptoms is the use of com-posite scores of disturbed sleeping or insomnia(8–12).These composite scores give a good overall notion of the association, but it is not possible to determine from these composites which specific sleep problem precedes anxiety and depression. This knowledge might facilitate the identification of children at risk for these disorders.

In a population-based prospective study,we examined whether sleep problems in infancy and early toddlerhood precede anxiety or depressive symptoms at the age of3years.The specific sleep problems and habits under study were dyssomnia, parasomnia,sleep duration,and rhythmicity.We also examine sleep-related parenting behavior(set bedtime,parental presence during sleep onset,co-sleeping),as these behaviors may interfere with the development of autonomous sleep habits(15–17).

METHODS

Design and Study Population

This study was embedded in Generation R,a population-based cohort from fetal life onward(18).All pregnant women(expected delivery date April 2002–January2006)living in Rotterdam,The Netherlands,were invited to participate by their midwife or obstetrician during routine visits.The partic-ipation rate was estimated at61%.

We obtained written informed consent from all participants.The Medical Ethical Committee of the Erasmus Medical Center,Rotterdam approved the study.

Data on sleep problems at2months or24months were available for6,367 children.Those without information on child behavior at36months(n?

From the Departments of Child and Adolescent Psychiatry(P.W.J.,N.S.S., F.C.V.,H.T.)and Epidemiology(A.H.,V.W.V.J.,H.T.),The Generation R Study Group(N.S.S.,V.W.V.J.),Department of Paediatrics(V.W.V.J.),Eras-mus Medical Center-University Medical Center Rotterdam,Rotterdam,The Netherlands.

Address correspondence and reprint requests to Henning Tiemeier,MD, PhD,Erasmus Medical Center-University Medical Center Rotterdam,Depart-ment of Child and Adolescent Psychiatry,P.O.Box2060,3000CB Rotter-dam,The Netherlands.E-mail:h.tiemeier@erasmusmc.nl

Received for publication April14,2010;revision received November3, 2010.

This first phase of the Generation R Study is supported,in part,by Erasmus Medical Center-University Medical Center Rotterdam;Erasmus University Rotterdam;and the Netherlands Organization for Health Research and De-velopment(ZonMW“Geestkracht”program10.000.1003).The work was also funded,in part,by Grant602from the Sophia Foundation for Medical Research SSWO(P.W.J.)and VIDI Grant017.106.370from the Netherlands Organization for Scientific Research(NWO-ZonMW)(H.T.).

The authors have not disclosed any potential conflicts of interest.

DOI:10.1097/PSY.0b013e31820a4abb