You are on page 1of 11

Physical Activity, Sedentary

Behavior, and Symptoms of Major
Depression in Middle Childhood
Tonje Zahl, MSC,a,b Silje Steinsbekk, PhD,b Lars Wichstrøm, PhDa,b

OBJECTIVE: The prospective relation between physical activity and Diagnostic and Statistical abstract
Manual of Mental Disorders-defined major depression in middle childhood is unknown,
as is the stability of depression. We therefore aimed to (1) determine whether there are
reciprocal relations between moderate-to-vigorous physical activity (MVPA) and sedentary
behavior, on one hand, and Diagnostic and Statistical Manual of Mental Disorders, Fourth
Edition defined symptoms of major depressive disorder, on the other and (2) assess the
extent of stability in depressive symptoms from age 6 to 10 years.
METHODS: A community sample of children living in Trondheim, Norway, comprising a total
of 795 6-year-old children was followed up at 8 (n = 699) and 10 (n = 702) years of age.
Physical activity was recorded by accelerometry and symptoms of major depression were
measured through semistructured clinical interviews of parents and children. Bidirectional
relationships between MVPA, sedentary activity, and symptoms of depression were
analyzed through autoregressive cross-lagged models, and adjusted for symptoms of
comorbid psychiatric disorders and BMI.
RESULTS: At both age 6 and 8 years, higher MVPA predicted fewer symptoms of major
depressive disorders 2 years later. Sedentary behavior did not predict depression, and
depression predicted neither MVPA nor sedentary activity. The number of symptoms of
major depression declined from ages 6 to 8 years and evidenced modest continuity.
CONCLUSIONS: MVPA predicts fewer symptoms of major depression in middle childhood, and
increasing MVPA may serve as a complementary method to prevent and treat childhood
depression.

aNTNU Social Reseach, Trondheim, Norway; and bDepartment of Psychology, Norwegian University of Science WHAT’S KNOWN ON THIS SUBJECT: Moderate-
and Technology, Trondheim, Norway to-vigorous physical activity (MVPA) reduces the
likelihood of depression in adolescents and adults,
Ms Zahl carried out initial analyses and drafted initial manuscript; Dr Steinsbekk conceptualized,
designed, and supervised the study; Dr Wichstrøm obtained funding and conceptualized, and is cross-sectionally associated with depression
designed, and supervised the study; and all authors approved the final manuscript as submitted. in children. However, the prospective relation
between MVPA, sedentary time, and depression in
DOI: 10.1542/peds.2016-1711
childhood is unknown.
Accepted for publication Nov 8, 2016
WHAT THIS STUDY ADDS: Objectively measured
Address correspondence to Tonje Zahl, MSC, NTNU Social Research, Dragvoll Allé 38B, 7048
MVPA at ages 6 and 8 predict fewer symptoms
Trondheim, Norway. E-mail: tonje.zahl@samfunn.ntnu.no
of depression in children 2 years later, whereas
PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275). sedentary activity is prospectively unrelated to
Copyright © 2017 by the American Academy of Pediatrics depression. Depression does not forecast reduced
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant physical activity in children.
to this article to disclose.
FUNDING: This research was supported by grants 228685 and 213793 from the Research Council To cite: Zahl T, Steinsbekk S, Wichstrøm L. Physical Activity,
of Norway and grant FO5148 from the Norwegian Extra Foundation for Health and Rehabilitation. Sedentary Behavior, and Symptoms of Major Depression in
Middle Childhood. Pediatrics. 2017;139(2):e20161711

Downloaded from by guest on June 24, 2017
PEDIATRICS Volume 139, number 2, February 2017:e20161711 ARTICLE
Both observational and treatment studies have nearly exclusively interviews) and accelerometry-
studies1–3 indicate that physical measured depression by means recorded MVPA and sedentary
activity (PA)—and moderate-to- of rating scales; however, the activity in a large community sample
vigorous PA (MVPA), in particular— correspondence between rating of Norwegian children monitored
may reduce the likelihood of major scale scores and the results from biannually from the ages of 6 to 10
depressive disorder (MDD) and/or clinical diagnostic interviews are years controlling for BMI and other
reduce the symptoms of MDD only moderate.17 Second, studies psychiatric symptoms.
in adolescents4 and adults.5–7 It of the PA-depression relationship
has recently become increasingly in children and adolescents have
clear that depression can also mainly used indirect methods to METHODS
be present in young children.8,9 measure children’s PA (ie, parent
Participants, Recruitment, and
Preventive10 and treatment efforts11 report or self-report, which are
Procedure
for childhood depression are only only moderately associated with
modestly effective, which suggests objectively measured PA); in The Trondheim Early Secure
that alternative or complementary fact, PA in community children Study consists of children from
interventions must be sought. may be overestimated by using the 2003 and 2004 birth cohorts
MVPA might possibly serve as a indirect methods.18 Hence, to avoid in Trondheim, Norway (n = 3456)
strategy for preventing or reducing confounding due to reporting bias, that were recruited by invitation
childhood depression. Indeed, a the PA–depression relationship letter, together with the Strengths
meta-analysis of randomized and should be investigated by using and Difficulties Questionnaire (SDQ)
quasiexperimental studies12 among objective measures of PA. Third, 4–16 version.26 Written consent
children in late childhood and early sedentary activity and MVPA are to participate was obtained when
adolescence suggested a small short- negatively, but far from perfectly, attending ordinary community health
term effect of PA interventions. correlated.19 In other words, some checkups for their 4-year-olds and
Given the waxing and waning of children may be periodically highly the completed SDQ was delivered. As
depression,13 it is important also to active but may nonetheless spend can be seen in Fig 1, the vast majority
discern whether long-term effects are much time engaged in sedentary (n = 3358, 97.2%) of children who
present. Toward this end, prospective behavior. A substantial amount of were invited to participate appeared
community studies might prove research has focused on MVPA; at the city’s well-child clinics. The
valuable. Importantly, psychomotor however, several studies have SDQ total problem scores (20 items)
retardation is 1 symptom of MDD,14 revealed that sedentary behavior, were divided into 4 strata (cutoffs:
and depressed children show not necessarily MVPA, might predict 0 to 4, 5 to 8, 9 to 11, and 12 to
higher levels of motor inactivity depression.20,21 Thus, we will 40), where drawing probabilities
than controls,6,15 which might investigate the extent to which time increased with increasing SDQ
explain the association reported. spent in sedentary activities predicts scores (0.37, 0.48, 0.70, and 0.89
Longitudinal studies are therefore symptoms of MDD over and above in the 4 strata) to increase sample
needed to reveal whether depression the effects of MVPA. Fourth, there is variability. The PA measurements
predicts reduced PA and whether considerable comorbidity between were included beginning with the
reduced PA increases the risk for depression and other psychiatric second wave of the data collection
depression in children, as has been disorders in children,22,23 and (6 years) and onward; therefore, the
shown in adolescents and adults.5,6,16 these disorders might be related to data used in the present inquiry were
Moreover, to our knowledge, no PA, confounding the relationship taken from 6-year (2009 to 2011,
longitudinal study has examined between depression and PA as a n = 795), 8-year (2011 to 2013, n =
the relationship between PA and result. Fifth, although the evidence is 699), and 10-year (2013 to 2015,
depression in middle childhood by mixed,24 a bidirectional relationship n = 702) assessments. In all, 799
using a community based sample. between childhood depression children had usable data from at least
and BMI is indicated,25 and BMI 1 measurement, and thus constitute
To identify the potentially beneficial should therefore be controlled for. the analytical sample. Attrition was
effects of PA on depression (and vice To overcome these obstacles, we not selective according to the study
versa) in nonreferred children, we investigate the prospective reciprocal variables, except that more hours
examine the bidirectional relation relations between Diagnostic and in MVPA at age 8 predicted attrition
between MVPA and symptoms Statistical Manual of Mental Disorders, at age 10 (odds ratio = 2.38; 95%
of depression, while addressing Fourth Edition (DSM-IV) defined confidence interval [CI]: 1.22 to 4.83,
several important methodological symptoms of major depression P = .01); a bias that was adjusted
issues. First, previous observational (obtained by using diagnostic for in the analyses (see Statistical

Downloaded from by guest on June 24, 2017
2 ZAHL et al
processed by using accelerometer
analysis software (ActiGraph LLC,
Pensacola, FL).

Symptoms of MDD
The Preschool Age Psychiatric
Assessment (PAPA),31 a psychiatric
interview completed by parents using
a structured protocol, with both
mandatory and optional follow-up
questions, was used to assess MDD
symptoms at 6 years. A sum score
of DSM-IV defined MDD symptoms
constituted the outcome. At 8 and 10
years of age, the Child and Adolescent
Psychiatric Assessment (CAPA)32
was used. Both parents and children
were interviewed and a symptom
was considered to be present if it was
reported by either child or parent.
The PAPA and CAPA interviewers
(n = 9) had at least a Bachelor’s
degree in a relevant field and were
instructed by the developers of the
PAPA and CAPA. For the PAPA, 9%
of the interview audio recordings
were recoded by blinded raters, as
were 15% of the CAPA interviews.
The interrater reliability between
multiple pairs of raters was 0.90 for
symptoms of MDD in the PAPA and
0.83 in the CAPA.
FIGURE 1
Sample recruitment and follow-up.
Symptoms of Other Psychiatric
Disorders
Analysis). The study was approved minutes of activity per day were
Symptoms of anxiety (consisting
by the Regional Committee of included. Detailed information on
of number of symptoms of social
Medical and Health Research Ethics, accelerometer compliance is shown
phobia, separation anxiety,
Mid-Norway. in Supplemental Table 3. Because
generalized anxiety, and specific
young children’s activity is often
Measurements phobias), attention-deficit/
intermittent with short bursts, we hyperactivity (ADHD), oppositional
Physical Activity employed the commonly used 10 defiant (ODD), and conduct disorders
The children were instructed to wear seconds epoch length.27 We applied (CDs) were assessed following the
an ActiGraph GT3X accelerometer the Evenson et al29 cutoff point of same procedures as for symptoms
(Manufacturing Technology, Inc, ≥2296 cpm for MVPA because it of MDD. Only the parents were
Fort Walton Beach, FL) around their has shown superior classification interviewed by using the CAPA with
waist for 7 consecutive days, 24 ability in children across the ages respect to ADHD. The interrater
hours a day, and only remove it relevant to the current study.30 reliabilities for PAPA/CAPA ranged
when bathing or showering. Only Minutes per day with ≤100 cpm from 0.85 to 0.97.
daytime activity (06:00–23:59) was was considered sedentary activity,
included. Sequences of consecutive a cutoff that is widely used and has Body Mass Index
zero counts lasting ≥20 minutes excellent classification accuracy.30 In The children’s weight was measured
were interpreted as nonwear time.27,28 the analyses, MVPA and sedentary by using a digital scale (Tanita
Only those participants with ≥3 activity were represented in BC20MA), and the height was
days of recordings and ≥480 hours per day intervals. Data were assessed by using the Heightronic

Downloaded from by guest on June 24, 2017
PEDIATRICS Volume 139, number 2, February 2017 3
Digital Stadiometer (QuickMedical
Model 235A). Correction for indoor
clothing (0.5 kg) was applied. BMI
was calculated as kg/m2.

Statistical Analysis

Reciprocal relations between PA
and the symptoms of MDD were
examined in Mplus 7.31,33 using FIGURE 2
Autoregressive cross-lagged relations between MVPA, sedentary behavior, and major depressive
autoregressive cross-lagged analysis symptoms. MVPA and sedentary activity were represented in hours per day intervals. Correlations
within a structural equation among the variables within each time point and the nonsignificant paths are not shown to simplify
framework. The symptoms of MDD, the model illustration, as are the autoregressive paths from 6 to 10 years.
MVPA, and sedentary activity at age
10 were regressed on measures TABLE 1 Descriptive Statistics
of these variables and covariates Descriptives of Study Variables (Possible Range) N M SD
at age 8, whereas age 8 measures
1. Number of MDD symptoms 6 y (0 to 9) 793 0.52 0.73
were regressed on measures at age 2. Number of MDD symptoms 8 y (0 to 9) 697 0.46 0.79
6, as shown in Fig 2. To allow for 3. Number of MDD symptoms 10 y (0 to 9) 629 0.52 0.90
sleeper effects (ie, the effects from 4. MVPA 6 y (hours per day) 697 1.19 0.40
age 6 to age 10 measures bypassing 5. MVPA 8 y (hours per day) 607 1.18 0.43
6. MVPA 10 y (hours per day) 684 1.09 0.40
measures at age 8), autocorrelated
7. Sedentary 6 y (hours per day) 689 8.58 0.84
paths were allowed from ages 6 to 8. Sedentary 8 y (hours per day) 607 9.22 1.06
10. Measures obtained at the same 9. Sedentary 10 y (hours per day) 634 9.94 1.07
time point in time (ie, at ages 6, 8, 10. Anxiety 6 y (0 to 21) 793 0.87 1.52
and 10, respectively) were allowed 11. Anxiety 8 y (0 to 21) 697 0.89 1.25
12. ADHD 6 y (0 to 18) 793 1.30 2.24
to correlate. To examine overall
13. ADHD 8 y (0 to 18) 688 1.20 2.40
changes in level of MDD, sedentary 14. CD 6 y (0 to 9)a 793 0.22 0.50
activity and MVPA latent growth 15. CD 8 y (0 to 15) 697 0.30 0.60
curve analyses were conducted. 16. ODD 6 y (0 to 8) 793 0.96 0,47
Because the attrition analyses 17. ODD 8 y (0 to 8) 697 1.08 1.39
18. BMI 6 y 658 15.60 1.51
indicated that the data were missing
19. BMI 8 y 675 16.62 1.97
not at random (MNAR), missing a Six items in the CD scale were removed at age 6 because they were not considered age appropriate.
data were handled according
to a full information maximum
likelihood procedure. Because RESULTS years (Mgrowth= 0.32 [95% CI, 0.27
counts of depressive symptoms are The descriptive statistics are shown to 0.37]) and increased further from
right skewed,34 a robust maximum in Table 1. The rate of DSM-IV defined ages 8 to 10 years (Mgrowth= 0.36
likelihood estimator was used, which MDD diagnosis, ranged from 0.3% [95% CI, 0.30 to 0.42]). Regarding
is robust to moderate deviations (age 6) to 0.4% (age 8), underscoring rank order stability, Table 2 shows
from normality. As oversampling was the need, at this age, to analyze MDD that the symptoms for MDD and
applied on the basis of mental health continuously as symptom counts. sedentary activity were modestly
problems, the data were weighted Supplementary piecewise growth stable, whereas we found higher
back with a factor determined curve analyses revealed that MDD stability for MVPA.
as the number of children in the decreased from ages 6 to 8 years
population in each stratum divided (Mgrowth= −0.13 [95% CI, −0.17 to Cross-Sectional Findings
by the number of participants in −0.10]) but increased from ages 8 The symptoms of MDD were
each stratum to arrive at correct to 10 years (Mgrowth= 0.03 [95% CI, negatively correlated with MVPA at
population estimates. To examine 0.00 to 0.06]). Minutes of MVPA per ages 8 and 10, but were unrelated
gender differences in the estimates, a day did not change from ages 6 to 8 to sedentary activity (Table 2). As
Wald test was employed to compare years (Mgrowth= −0.17 [95% CI, −1.35 expected, the symptoms of MDD
the fit of 2 models, one in which the to 1.01]) but decreased from ages 8 covaried with the symptoms of other
path at hand was fixed as identical for to 10 years (Mgrowth= −2.62 [95% CI, disorders at all time points. The
the 2 genders and the other in which −3.75 to −1.49]). Finally, sedentary symptoms of these other disorders
the path was freely estimated.26 activity increased from ages 6 to 8 were generally unrelated to MVPA

Downloaded from by guest on June 24, 2017
4 ZAHL et al
−0.17***
−0.10*
and sedentary activity, although we

0.84***
−0.08

−0.03

−0.02

−0.03

−0.01

−0.04
0.05
0.05
0.03

0.04

0.26

0.07
0.02
0.06
0.01

1
found higher MVPA among 6-year-old
children who had more symptoms of

−0.13**
−0.09*
−0.06

−0.01

−0.03

−0.04

−0.01
anxiety disorders and less sedentary
0.06
0.03
0.05

0.00

0.05

0.01

0.06

0.03
0.04

0.00


1
activity among those 6-year-old
children who had elevated symptoms
0.21***
0.27***

0.17***

0.35***

0.30***
0.35***
0.16***
0.33***
0.35***
−0.05

−0.04
0.16*

0.10*

of ADHD.
0.07

0.04
0.04



1
19

Prospective Associations
−0.11**
0.44***
0.21***
0.19***

0.36***

0.27***

0.38***
0.30***
0.34***
0.23***
−0.02

−0.05
0.09*
0.04
0.06




1
The main results from the
autoregressive cross-lagged
0.19***

0.22***

0.27***
0.29***
0.18***
−0.01

−0.04

−0.06
0.12**

0.14*
0.08*

0.08*

model examining the bidirectional
0.06
0.06





1
relationships between MVPA,
sedentary activity, and major
0.17***

0.18***
−0.04

0.12**
0.10*

0.09*

0.09*
0.05
0.04
0.03
0.04

0.03

0.01

depression are shown in Fig 2.





1
For complete model results, see
18

Supplemental Table 4.
−0.09*
0.25***
0.29***
0.19***

0.28***

0.47***

0.58***
−0.01

−0.06

−0.09
0.07
0.06







1
17

The model did fit the data well (χ2 =
21.76, df = 24, P = .59, comparative
−0.10**
0.35***
0.20***
0.14***

0.33***

0.35***
−0.04

−0.07

−0.07
0.10**
16

0.08

fit index = 1.00, Tucker-Lewis index =







1

1.007, root mean square error of
15

approximation = 0.000). At both 6
0.23***
0.47***
0.21***

0.32***
−0.01
−0.01
−0.06
−0.07

−0.07

0.01









1

and 8 years, higher levels of MVPA
14

predicted fewer symptoms of MDD 2
years later. The reduction was ∼0.20
0.48***
0.19***

−0.06
−0.03
−0.07
−0.06

−0.00
0.12*

0.01









1
13

symptoms of depression per daily
hour spent in MVPA. The effect of
−0.17***
−0.47***
−0.16**

−0.22**
12

0.28***
−0.00

MVPA on depression from age 8 to
0.06

0.06









1

10 was seemingly stronger than the
11

effect from age 6 to 8. Standardized
−0.42***
−0.15***
−0.10*

0.30***
−0.03

coefficients revealed effect sizes of β
0.00

0.03









1

−0.08 (at age 6 to 8) and β −0.11 at
10

age 8 to 10), respectively. However,
−0.40***
−0.18***
−0.12**
−0.04
0.02
0.03








when comparing a model in which
1
9

these 2 path coefficients were fixed
as equal with a model in which they
−0.17***
0.37***
0.47***
−0.06
−0.03
8









1

were freely estimated, no significant
difference was found (Wald = 0.41,
7

−0.16**

df = 1, P = .41). A similar procedure
0.40***
-0.10**
TABLE 2 Bivariate Correlations Between Study Variables

0.02










1

was used to test for the gender-
6

specific effects of MVPA on later
−0.08**
−0.01

−0.07

symptoms of MDD, and no such










1
5

differences were found (6 to 8 years:
Wald = 0.47, df = 1, P = .49; 8 to 10
0.19***
0.25***
4

years: Wald = 0.34, df = 1, P = .56).











1

Because MDD symptoms are heavily
3

right-skewed with many children
0.27***













1

evincing no symptoms and just a
2

few evincing many symptoms, there












is a possibility that the obtained
1
1

—, not applicable.

effect of MVPA on MDD was due to
7. Sedentary

8. Sedentary

9. Sedentary
6. MVPA 10 y

12. ADHD 6 y
13. ADHD 8 y
3. MDD 10 y
4. MVPA 6 y
5. MVPA 8 y

16. ODD 6 y
17. ODD 8 y
18. BMI 6 y
19. BMI 8 y
10. Anxiety

11. Anxiety
1. MDD 6 y
2. MDD 8 y

skewness and count nature of MDD
*** P < .001.
14. CD 6 y
15. CD 8 y

** P < .01.
* P < .05.
10 y

symptoms. We therefore run the path
6y

8y

6y

8y

analysis treating MDD symptoms as

Downloaded from by guest on June 24, 2017
PEDIATRICS Volume 139, number 2, February 2017 5
having a negative binominal (count) symptoms of MDD, 2 potential adolescents and adults has suggested
distribution. MVPA still predicted mechanisms related to PA’s 2 not that sedentary behavior may increase
reduced numbers of MDD symptoms mutually exclusive components depression.20 It is thus likely that the
from age 6 to 8 (B = −0.58 [95% (the activity and the physical effect of sedentariness on depression
CI: −0.95 to −0.21], P = .001) and components), can be hypothesized. is age-dependent. However, because
from age 8 to 10 (B = −0.58 [95% CI: Treatment studies applying activity psychomotor retardation and loss
−1.03 to −0.13], P = .002). Moreover, scheduling have indicated that of energy are among the symptoms
depression did not predict later increasing activity, not necessarily of MDD, it is essential to adjust for
MVPA, and no effects of sedentary just PA, may reduce depression.38 previous MDD when examining
activity on depression, or vice versa, Several explanations for this MDD- sedentary activity’s contribution to
were detected. reducing effect have been proposed: MDD, which has been investigated
(1) Engaging in activities may distract only to a limited extent.21,49 Further,
DISCUSSION from ruminating over negative in many of these prospective
events, and ruminations may worsen studies, self-reports of inactivity
To aid the preventive and treatment or prolong depression.39,40 (2) A and depression have been applied
efforts of depression among children, substantial part of MVPA in children that may inflate their relationship
we examined the bidirectional consists of play or sports activities,41 on a prospective basis, due to the
relations between MVPA, sedentary and playing or engaging in sports common methods that are employed.
activity, and MDD symptoms over may bolster self-efficacy and self- Moreover, it is essential to adjust
3 waves of data collection in a large esteem in children, which has been for lack of MVPA when estimating
community sample of children. suggested to prevent depression.42,43 the effects of sedentariness because
Higher levels of MVPA at 6 and 8, (3) Finally, when children are MVPA and sedentary activity are
respectively, predicted fewer MDD physically active, they tend to be negatively correlated. In sum, the lack
symptoms 2 years later. No effects of with other children.41 Although peer- of comparable studies underscores
sedentary activity were found for MDD rejections and bullying also occur in the need to replicate our findings.
symptoms, and MDD symptoms did sports,44 physically active children
not predict later PA (or lack thereof). may be more socially integrated in Levels of PA and Symptoms of MDD
peer groups than less active children. During Middle Childhood
MVPA Predicts Fewer Depressive
Symptoms Such peer acceptance, which is The level of symptoms of MDD
likely to result in social support, is a was stable from 6 to 10 years.
The identified effects of PA on
potential buffer against depression.45 MDD symptoms evidenced some
depression extend findings from homotypical continuity, which
observational studies of adolescents Regarding the physical component extends the findings from short-term
and adults35 by documenting that of PA, several physiologic and longitudinal studies on MDD in young
this relationship is also present in biochemical mechanisms have been children.9,50 The observed reduction
middle childhood. We further add proposed, such as the demonstration in MVPA and increase in sedentary
to existing knowledge by showing of PA leading to higher availability of behavior are consistent with findings
that predictive effects are present neurotransmitters, which is assumed from previous research.19
when we examine interview-based to have antidepressant effects.46,47 In
DSM-defined symptoms of MDD addition, regular PA has a favorable Limitations
rather than questionnaires, and impact on neuronal functions and Although individuals fulfilling the DSM
when applying objectively measured structure along with increased cutoff of 5 or more MDD symptoms
PA, and account for the potential cognitive functions.48 do not seem qualitatively different
effects of other psychiatric disorders
from those with just under cutoffs,51
and BMI. Although the effects of No Prospective Relation Between and even though the correlates and
MVPA were small, they are similar Sedentary Activity and Symptoms of predictors of subclinical depression
to those obtained by psychosocial MDD
are similar to those of the disorder,52
intervention programs in children36
Our study is the first to objectively our findings do not necessarily
and adolescents.37 Along with the
examine sedentary activity and generalize to the disorder itself. Thus,
fact that nearly all children can be
later depression in early and research with substantially larger
targeted in efforts to increase MVPA,
middle childhood, and we find samples is required to determine
the gains at the population level
no prospective relation between whether objectively measured MVPA
might be substantial.
sedentary behavior and symptoms would decrease the risk of MDD in
Although the current study did not of MDD. Although there are some community children. Nonetheless,
address why PA may reduce future important exceptions,49 research on because previous research has

Downloaded from by guest on June 24, 2017
6 ZAHL et al
shown that there is a continuity in enhancement58 or increased social
interactions59 may explain at least
ABBREVIATIONS
depressive symptoms from childhood
to adolescence and later adulthood,53 some of the results. The prevalence of ADHD: attention-deficit/
that an elevated level of MDD psychiatric disorders is generally low hyperactivity disorder
symptoms increases the risk of in Norway,60 but whether the impact CAPA: Child and Adolescent
later MDD,54,55 and that subclinical of MVPA on depressive symptoms also Psychiatric Assessment
depression may entail substantial differ between countries should be CD: conduct disorder
impairment (also in the long run),56 examined in future studies. CI: confidence interval
our findings suggest that increasing DSM-IV: Diagnostic and
MVPA at the population level may lead Statistical Manual of
to reduced symptoms of depression— CONCLUSIONS Mental Disorders, Fourth
and the impairment that accompanies Edition
MVPA predicts fewer future MDD
these symptoms in some children. MDD: major depressive
symptoms in middle childhood,
Second, data were MNAR, which may disorder
and such symptoms are moderately
MNAR: missing not at random
have led to biased results. However, stable from the ages of 6 to 10 years.
MVPA: moderate-to-vigorous
the selectivity of this attrition was Sedentary activity in children does
physical activity
modest and we used all available not alter the risk of future symptoms
PA: physical activity
data in a full information maximum of depression, and depression does
PAPA: Preschool Age Psychiatric
likelihood procedure, which leads to not influence the likelihood of MVPA
Assessment
less biased results than complete case or inactivity. Although the effect
ODD: oppositional defiant
analysis when the data are MNAR.57 was small, our results indicate that
disorder
Nonetheless, we cannot exclude that increasing MVPA in children at
SDQ: Strengths and Difficulties
such bias along with unmeasured the population level may prevent
Questionnaire
confounders such as self-image depression, at least at subclinical levels.

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

REFERENCES
1. Silveira H, Moraes H, Oliveira N, 5. Pereira SMP, Geoffroy MC, Power from preschool to first grade.
Coutinho ESF, Laks J, Deslandes C. Investigation of bidirectional Dev Psychopathol. 2016;28(4 pt
A. Physical exercise and clinically associations between depressive 2):1517–1530
depressed patients: a systematic symptoms and physical activity from 10. Beardslee WR, Gladstone TRG.
review and meta-analysis. age 11 to 50 years in the 1958 British Prevention of childhood depression:
Neuropsychobiology. 2013;67(2): Birth Cohort. Lancet. 2013;382:80 recent findings and future prospects.
61–68 Biol Psychiatry. 2001;49(12):1101–1110
6. Azevedo Da Silva M, Singh-Manoux
2. Wegner M, Helmich I, Machado S, Nardi A, Brunner E, et al. Bidirectional 11. Michael KD, Crowley SL. How
AE, Arias-Carrion O, Budde H. Effects association between physical effective are treatments for child
of exercise on anxiety and depression activity and symptoms of anxiety and and adolescent depression? A meta-
disorders: review of meta-analyses depression: the Whitehall II study. Eur J analytic review. Clin Psychol Rev.
and neurobiological mechanisms. Epidemiol. 2012;27(7):537–546 2002;22(2):247–269
CNS Neurol Disord Drug Targets. 7. Pinto Pereira SM, Geoffroy MC, 12. Brown HE, Pearson N, Braithwaite RE,
2014;13(6):1002–1014 Power C. Depressive symptoms and Brown WJ, Biddle SJ. Physical activity
3. Josefsson T, Lindwall M, Archer T. physical activity during 3 decades in interventions and depression in
Physical exercise intervention in adult life: bidirectional associations children and adolescents: a systematic
depressive disorders: meta-analysis in a prospective cohort study. JAMA review and meta-analysis. Sports Med.
and systematic review. Scand J Med Psychiatry. 2014;71(12):1373–1380 2013;43(3):195–206
Sci Sports. 2014;24(2):259–272 8. Luby JL, Heffelfinger AK, Mrakotsky C, 13. Birmaher B, Ryan ND, Williamson
et al. The clinical picture of depression DE, et al. Childhood and adolescent
4. Jerstad SJ, Boutelle KN, Ness KK,
in preschool children. J Am Acad Child depression: a review of the past 10
Stice E. Prospective reciprocal
Adolesc Psychiatry. 2003;42(3):340–348 years. Part I. J Am Acad Child Adolesc
relations between physical
activity and depression in female 9. Reinfjell T, Kårstad SB, Berg-Nielsen Psychiatry. 1996;35(11):1427–1439
adolescents. J Consult Clin Psychol. TS, Luby JL, Wichstrøm L. Predictors 14. Birmaher B, Brent D. Practice
2010;78(2):268–272 of change in depressive symptoms parameter for the assessment and

Downloaded from by guest on June 24, 2017
PEDIATRICS Volume 139, number 2, February 2017 7
treatment of children and adolescents 24. Muhlig Y, Antel J, Focker M, 35. Teychenne M, Ball K, Salmon J. Physical
with depressive disorders. J Am Hebebrand J. Are bidirectional activity and likelihood of depression
Acad Child Adolesc Psychiatry. associations of obesity and depression in adults: a review. Prev Med.
2007;46(11):1503–1526 already apparent in childhood and 2008;46(5):397–411
adolescence as based on high-quality 36. Horowitz JL, Garber J. The prevention
15. Aronen ET, Simola P, Soininen M. Motor
studies? A systematic review. Obesity of depressive symptoms in children
activity in depressed children. J Affect
Rev . 2016;17(3):235–249 and adolescents: A meta-analytic
Disorder. 2011;133(1–2):188–196
25. Luppino FS, de Wit LM, Bouvy PF, et al. review. J Consult Clin Psychol.
16. Epstein LH, Paluch RA, Coleman KJ,
Overweight, obesity, and depression: a 2006;74(3):401–415
Vito D, Anderson K. Determinants of
systematic review and meta-analysis 37. Stice E, Shaw H, Bohon C, Marti CN,
physical activity in obese children
of longitudinal studies. Arch Gen Rohde P. A meta-analytic review of
assessed by accelerometer and
Psychiatry. 2010;67(3):220–229 depression prevention programs for
self-report. Med Sci Sport Excer.
1996;28(9):1157–1164 26. Goodman R. The strengths and children and adolescents: factors that
difficulties questionnaire: a research predict magnitude of intervention
17. Sveen TH, Berg-Nielsen TS, Lydersen
note. J Child Psychol Psychiatry. effects. J Consult Clin Psychol.
S, Wichstrøm L. Detecting psychiatric
1997;38(5):581–586 2009;77(3):486–503
disorders in preschoolers: screening
with the strengths and difficulties 27. Cain KL, Sallis JF, Conway TL, Van Dyck 38. Cuijpers P, van Straten A, Warmerdam
questionnaire. J Am Acad Child Adolesc D, Calhoon L. Using accelerometers L. Behavioral activation treatments
Psychiatry. 2013;52(7):728–736 in youth physical activity studies: a of depression: a meta-analysis. Clin
review of methods. J Phys Act Health. Psychol Rev. 2007;27(3):318–326
18. Adamo KB, Prince SA, Tricco AC,
2013;10(3):437–450 39. Rood L, Roelofs J, Bögels SM, Nolen-
Connor-Gorber S, Tremblay M. A
comparison of indirect versus direct 28. Ward DS, Evenson KR, Vaughn A, Hoeksema S, Schouten E. The influence
measures for assessing physical Rodgers AB, Troiano RP. Accelerometer of emotion-focused rumination and
activity in the pediatric population: a use in physical activity: best practices distraction on depressive symptoms
systematic review. Int J Pediatr Obes. and research recommendations. in non-clinical youth: a meta-
2009;4(1):2–27 Med Sci Sports Exerc. 2005;37(suppl analytic review. Clin Psychol Rev.
11):S582–S588 2009;29(7):607–616
19. Basterfield L, Adamson AJ, Frary JK,
et al. Longitudinal study of physical 29. Evenson KR, Catellier DJ, Gill K, Ondrak 40. Nolen-Hoeksema S. Responses to
activity and sedentary behavior in KS, McMurray RG. Calibration of depression and their effects on the
children. Pediatrics. 2011;127(1). two objective measures of physical duration of depressive episodes. J
Available at: www.pediatrics.org/cgi/ activity for children. J Sports Sci. Abnorm Psychol. 1991;100(4):569–582
content/full/127/1/e24 2008;26(14):1557–1565 41. Macdonald-Wallis K, Jago R, Sterne
20. Page AS, Cooper AR, Griew P, Jago R. 30. Trost SG, Loprinzi PD, Moore R, Pfeiffer JAC. Social network analysis of
Children’s screen viewing is related to KA. Comparison of accelerometer cut childhood and youth physical activity:
psychological difficulties irrespective points for predicting activity intensity a systematic review. Am J Prev Med.
of physical activity. Pediatrics. in youth. Med Sci Sports Exerc. 2012;43(6):636–642
2010;126(5). Available at: www. 2011;43(7):1360–1368 42. Sowislo JF, Orth U. Does low
pediatrics.org/cgi/content/full/126/5/ 31. Egger HL, Erkanli A, Keeler G, Potts self-esteem predict depression
e1011 E, Walter BK, Angold A. Test-Retest and anxiety? A meta-analysis of
21. Sund AM, Larsson B, Wichstrøm L. Role Reliability of the Preschool Age longitudinal studies. Psychol Bull.
of physical and sedentary activities Psychiatric Assessment (PAPA). J 2013;139(1):213–240
in the development of depressive Am Acad Child Adolesc Psychiatry. 43. Ekeland E, Heian F, Hagen Kåre B,
symptoms in early adolescence. 2006;45(5):538–549 Abbott Jo M, Nordheim L. Exercise to
Soc Psychiatry Psychiatr Epidemiol. 32. Angold A, Costello EJ. The Child and improve self-esteem in children and
2011;46(5):431–441 Adolescent Psychiatric Assessment young people. Cochrane Database Sys
22. Keiley M, Bates J, Dodge K, Pettit G. (CAPA). J Am Acad Child Adolesc Rev. 2004(1):CD003683
A cross-domain growth analysis: Psychiatry. 2000;39(1):39–48 44. Stirling AE, Bridges EJ, Cruz EL,
externalizing and internalizing 33. Muthén LK, Muthén BO. Mplus User’s Mountjoy ML; Canadian Academy of
behaviors during 8 years of Guide. 7th ed. Los Angeles, CA: Muthén Sport and Exercise Medicine. Canadian
childhood. J Abnormal Child Psychol. & Muthén; 1998–2015 Academy of Sport and Exercise
2000;28(2):161–179 Medicine position paper: abuse,
34. Benson J, Fleishman J. The
23. Lilienfeld S. Comorbidity between and robustness of maximum likelihood harassment, and bullying in sport. Clin
within childhood externalizing and and distribution-free estimators J Sport Med. 2011;21(5):385–391
internalizing disorders: reflections and to non-normality in confirmatory 45. Epkins C, Heckler D. Integrating
directions. J Abnormal Child Psychol. factor analysis. Qual Quant. etiological models of social anxiety
2003;31(3):285–291 1994;28(2):117–136 and depression in youth: evidence

Downloaded from by guest on June 24, 2017
8 ZAHL et al
for a cumulative interpersonal risk meta-analytic review. Acta Psychiatr 56. Cuijpers P, Smit F. Subthreshold
model. Clin Child Fam Psychol Rev. Scand. 2007;115(6):434–441 depression as a risk indicator
2011;14(4):329–376 52. Shankman SA, Lewinsohn PM, Klein for major depressive disorder: a
DN, Small JW, Seeley JR, Altman SE. systematic review of prospective
46. Krishnan V, Nestler EJ. The molecular
Subthreshold conditions as precursors studies. Acta Psychiatr Scand.
neurobiology of depression. Nature.
for full syndrome disorders: a 2004;109(5):325–331
2008;455(7215):894–902
15-year longitudinal study of multiple 57. Schafer JL, Graham JW. Missing data:
47. Craft LL, Perna FM. The benefits of
diagnostic classes. J Child Psychol our view of the state of the art. Psychol
exercise for the clinically depressed.
Psychiatry. 2009;50(12):1485–1494 Methods. 2002;7(2):147–177
Primary Care Companion J Clin
Psychiatry. 2004;6(3):104–111 53. Dekker MC, Ferdinand RF, van Lang ND, 58. Kirkcaldy BD, Shephard RJ, Siefen RG.
Bongers IL, van der Ende J, Verhulst The relationship between physical
48. Chaddock-Heyman L, Erickson KI,
FC. Developmental trajectories of activity and self-image and problem
Holtrop JL, et al Aerobic fitness is
depressive symptoms from early behaviour among adolescents. Soc
associated with greater white matter
childhood to late adolescence: Psychiatry Psychiatr Epidemiol.
integrity in children. Frontiers Human
gender differences and adult 2002;37(11):544–550
Neurosci. 2014;8:584
outcome. J Child Psychol Psychiatry. 59. Monshouwer K, Ten Have M, van
49. Zhai L, Zhang Y, Zhang D. Sedentary 2007;48(7):657–666 Poppel M, Kemper H, Vollebergh W.
behaviour and the risk of depression:
54. Luby JL, Gaffrey MS, Tillman R, April LM, Possible mechanisms explaining the
a meta-analysis. Br J Sports Med.
Belden AC. Trajectories of preschool association between physical activity
2015;49(11):705–709
disorders to full DSM depression at and mental health: findings from the
50. Bufferd SJ, Dougherty LR, Carlson school age and early adolescence: 2001 Dutch Health Behaviour in School-
GA, Rose S, Klein DN. Psychiatric continuity of preschool depression. Am Aged Children Survey. Clin Psychol Sci.
disorders in preschoolers: continuity J Psychiatry. 2014;171(7):768–776 2013;1(1):67–74
from ages 3 to 6. Am J Psychiatry. 55. Pine DS, Cohen E, Cohen P, Brook J. 60. Wichstrøm L, Berg-Nielsen TS, Angold
2012;169(11):1157–1164 Adolescent depressive symptoms A, Egger HL, Solheim E, Sveen TH.
51. Cuijpers P, Smit F, van Straten as predictors of adult depression: Prevalence of psychiatric disorders
A. Psychological treatments moodiness or mood disorder? Am J in preschoolers. J Child Psychol
of subthreshold depression: a Psychiatry. 1999;156(1):133–135 Psychiatry. 2012;53(6):695–705

Downloaded from by guest on June 24, 2017
PEDIATRICS Volume 139, number 2, February 2017 9
Physical Activity, Sedentary Behavior, and Symptoms of Major Depression in
Middle Childhood
Tonje Zahl, Silje Steinsbekk and Lars Wichstrøm
Pediatrics 2017;139;; originally published online January 9, 2017;
DOI: 10.1542/peds.2016-1711
Updated Information & including high resolution figures, can be found at:
Services /content/139/2/e20161711.full.html
Supplementary Material Supplementary material can be found at:
/content/suppl/2017/01/05/peds.2016-1711.DCSupplemental.
html
References This article cites 57 articles, 2 of which can be accessed free
at:
/content/139/2/e20161711.full.html#ref-list-1
Subspecialty Collections This article, along with others on similar topics, appears in
the following collection(s):
Psychiatry/Psychology
/cgi/collection/psychiatry_psychology_sub
Permissions & Licensing Information about reproducing this article in parts (figures,
tables) or in its entirety can be found online at:
/site/misc/Permissions.xhtml
Reprints Information about ordering reprints can be found online:
/site/misc/reprints.xhtml

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly
publication, it has been published continuously since 1948. PEDIATRICS is owned, published,
and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk
Grove Village, Illinois, 60007. Copyright © 2017 by the American Academy of Pediatrics. All
rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.

Downloaded from by guest on June 24, 2017
Physical Activity, Sedentary Behavior, and Symptoms of Major Depression in
Middle Childhood
Tonje Zahl, Silje Steinsbekk and Lars Wichstrøm
Pediatrics 2017;139;; originally published online January 9, 2017;
DOI: 10.1542/peds.2016-1711

The online version of this article, along with updated information and services, is
located on the World Wide Web at:
/content/139/2/e20161711.full.html

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly
publication, it has been published continuously since 1948. PEDIATRICS is owned,
published, and trademarked by the American Academy of Pediatrics, 141 Northwest Point
Boulevard, Elk Grove Village, Illinois, 60007. Copyright © 2017 by the American Academy
of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.

Downloaded from by guest on June 24, 2017