Parents influence children

Literature Review

Parenting Practices and Children’s Physical Activity: An Integrative Review

Amy Hutchens, PhD(c), RN, CNE1, and Rebecca E. Lee, PhD2

Abstract

The purpose of this integrative review was to analyze the state of science concerning the influence of parenting practices on children’s physical activity (PA) levels. A total of 38 studies met the inclusion criteria after full-text review. The body of research is limited in experimental designs with only three studies measuring the influence of the intervention on parenting practices. Seven of the 30 quantitative studies (23.3%) found significant associations between parental role modeling of PA and children’s PA levels. Seven of the eight (87.5%) qualitative studies identified parental role modeling of PA as important in promoting children’s PA. Sixteen of the 30 (53.3%) quantitative studies found that parental support of PA was significantly associated with children’s PA. Five of the eight (62.5%) qualitative studies identified logistic support as supporting PA in children. The science could be expanded by the development of randomized controlled trials aimed at this area.

Keywords

parenting, exercise, child, review, school nursing

A dramatic reduction in the amount of physical activity (PA)

occurs between the school age and adolescent years. People

who are physically active typically live longer than those

who are not (Paffenbarger, Hyde, Wing, & Hsieh, 1986).

They also have lower rates of heart disease, stroke, type 2

diabetes, depression, and some cancers (American Heart

Association, 2013). This is also true for children. PA leads

to substantial physical and psychological health benefits

for youth (Strong et al., 2005). Parents influence children

PA levels; however, the extent and nature of this relation-

ship is not clearly defined or documented. Parenting prac-

tices include ways in which parents may shape their

children’s PA habits including increasing knowledge, mod-

eling PA, and providing accessibility to healthy PA options

(Davison & Campbell, 2005). It is critical to understand the

role that parents play in children’s activity behaviors. The

purpose of this integrative review is to analyze the state of

science concerning the influence of parenting practices on

children’s PA.

Purpose

PA is a vital component for good health and long life. It is

associated with increased fitness, improved cardiovascular

and metabolic disease risk factors, decreased likelihood of

developing type 2 diabetes, improved bone strength, and

improved mental well-being (National Physical Activity

Plan, 2014). In the United States, although 42% of children ages 6–11 years meet the recommended daily levels of 60 or

more min of moderate to vigorous PA per day, only 8% of youth aged 12- to 15 years old meet that recommendation

(National Physical Activity Plan, 2014). PA helps one main-

tain a healthy body weight (McTiernan et al., 2007). By

2030, 13 states could have obesity rates above 60% (Levi, Segal, St. Laurent, Lang, & Rayburn, 2012). Improvement in

PA levels may help to prevent obesity.

Although many factors have been shown to contribute to

children’s PA, parents have been identified as key in the

promotion of PA (O’Connor, Jago, & Barnowski, 2009).

Parents serve as important guides and role models for PA,

and they organize and fund children’s involvement in phys-

ical activities (Davison, Cutting, & Burch, 2003). Evidence

aimed at understanding the influence of parenting practices

on children’s PA levels is emerging, but the depth and

breadth are not clear. A review of parenting practices was

conducted by Xu, Wen, and Rissel (2015) but did not include

qualitative evidence (Xu, Wen, & Rissel, 2015), limiting the

depth of the review and conclusions that might be drawn

1 Fran and Earl Ziegler College of Nursing, University of Oklahoma Health

Sciences Center, Oklahoma City, OK, USA 2 Center for Health Promotion and Disease Prevention, College of Nursing

and Health Innovation, Arizona State University, Phoenix, AZ, USA

Corresponding Author:

Amy Hutchens, PhD(c), RN, CNE, Fran and Earl Ziegler College of Nursing,

University of Oklahoma Health Sciences Center, 1100 N. Stonewall,

Oklahoma City, OK 73117, USA.

Email: amy-hutchens@ouhsc.edu

The Journal of School Nursing 2018, Vol. 34(1) 68-85 ª The Author(s) 2017 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/1059840517714852 journals.sagepub.com/home/jsn

 

 

from it. That review found that parental encouragement and

support increase children’s PA (Xu et al., 2015). An inte-

grative review is necessary to provide unique insight through

the consideration of qualitative as well as quantitative evi-

dence. This review will critically analyze the state of science

concerning the influence of parenting practices and chil-

dren’s PA levels.

The family ecological model (FEM) and the ecological

model of physical activity (EMPA) provided the theoretical

background for the study (Davison, Jurkowski, & Lawson,

2012; Spence & Lee, 2003). The FEM was developed to

explain the relationships among the environment and parent-

ing aimed at the promotion of healthy lifestyles (Davison

et al., 2012). The FEM accounts for contextual and family

factors influencing children’s PA-related behaviors

(Davison et al., 2012). Parenting is a well-developed, main

component of the FEM. The FEM posits that to understand

behavior, one must consider the context in which the person

is living and that family provides the environment in which

behavior patterns emerge (Davison & Campbell, 2005). Par-

enting is encompassed within the EMPA as a dynamic envi-

ronmental linkage that helps to promote PA in children

(Spence & Lee, 2003). The model notes the direct and indi-

rect effects of multiple levels of the environment and their

influence on PA, and it accounts for the variations in distal

and proximal influences on PA. Distal influences of PA are

buffered by proximal factors (Spence & Lee, 2003). For

example, poor neighborhood conditions would have less

influence on children’s PA if parents were able to provide

a supportive home environment for PA.

The existing research suggests that parents influence chil-

dren’s PA by providing the context through behavior patterns

emerge (Davison & Campbell, 2005). The purpose of this

integrative review was to identify the PA parenting practices

that potentially influence children’s PA. The integrative

review is the broadest type of review; it allows for inclusion

of experimental and nonexperimental research to obtain a bet-

ter understanding of the phenomenon of concern (Whittemore

& Knafl, 2005) including problem identification, literature

search, data reduction, data display, and data comparison as

strategies to enhance rigor in integrative reviews.

Method

Eligibility Criteria

This review covered a 19-year period (1998–2017). A 19-

year period was used, as the combination of search terms did

not generate articles prior to 1998. The CINAHL, Ovid

Medline, and PubMed databases were searched in January

of 2017 for eligible studies. The search terms consisted of

combinations of “parenting,” “children,” “and “physical

activity.” Inclusion criteria for this study necessitated that

studies used qualitative or quantitative designs, were pub-

lished in English-language peer-reviewed journals, and

investigated parenting practices that influence PA in chil-

dren and adolescents. Reference lists of included papers

were examined to determine whether any of the studies met

the inclusion criteria. Studies were not considered if they did

not address parenting practices specific to PA. Studies solely

examining parenting style (without parenting practices)

were excluded. Parenting style as described by Maccoby and

Martin (1983) includes authoritarian, authoritative, permis-

sive, and uninvolved. Unpublished studies were not included

in the review. Four hundred and seven studies were initially

identified from the database searches. A total of 38 studies

met the inclusion criteria after full-text review (see flow-

chart in Figure 1).

Polit and Beck’s (2012) evidence hierarchy was used to

identify the levels of evidence for each study to determine

the strength of available evidence. The strongest level of

research ranked as Level 1 by Polit and Beck includes sys-

tematic reviews of randomized and nonrandomized control

trials. Level 2 includes single-randomized and nonrando-

mized clinical trials. The third level is systematic reviews

of correlational and observational studies. Level 4 incorpo-

rates single correlational and observational studies, and

Level 5 includes systematic reviews of descriptive and qua-

litative studies. The sixth level described by Polit and Beck

are single descriptive and qualitative studies. Last, the

seventh level entails opinions from experts.

Quality Assessment

Quantitative studies were evaluated for quality using the

Grading of Recommendations, Assessment, Development,

and Evaluation (GRADE) approach (Higgins & Green,

2011; see Table 1). This approach ranks the quality of evi-

dence as high, moderate, low, and very low. Rankings are

based on underlying methodology, and reviewers may mod-

ify the quality of a study based on an assessment of factors

that may increase or decrease the quality of the study. Ran-

domized control trials (RCTs) are ranked as high but can be

downgraded to moderate low or very low if they have lim-

itations in design, inconsistent results, imprecision of results,

or publication bias (Higgins & Green, 2011). Observational

studies are ranked as low but can be upgraded for reasons

that include a large magnitude effect. Observational studies

can also be downgraded to very low using the same criteria

used to downgrade RCTs.

A variety of methods were utilized for data collection

within the qualitative studies. Three of the studies employed

1:1 interviews for data collection (Bentley et al., 2012;

Hosseini, Anoosheh, Abbaszadeh, & Ehsani, 2013; Ickes,

Mahoney, Roberts, & Dolan, 2016). Three employed focus

groups (De Lepeleere, De Smet, Verloigne, Cardon, & De

Bourdeaudhuij, 2013; Hesketh, Hinkley, & Campbell, 2012;

Wright, Wilson, Griffin, & Evans, 2008). Nominal group

technique was used by two of the qualitative studies

(O’Connor, Cerin, et al., 2013; Suen, Cerin, & Wu, 2015).

Hutchens and Lee 69

 

 

Nominal group technique eliminates potential bias that may

occur in traditional focus groups by utilizing a structured

method, so that on person cannot dominate the discussion

(O’Connor, Cerin, et al., 2013), Only two of the qualitative

studies provided interrater reliability between coders. Those

were reported at r ¼ .87 and r ¼ .84 (De Lepeleere et al., 2013; Wright et al., 2008).

Data Extraction and Synthesis

Key concepts were identified, and methodological evidence

was evaluated (Whittemore & Knafl, 2005). The design of

this review also included principles of cross-case analysis

described by Miles, Huberman, and Saldana (2014). The

cross-case analysis seeks to enhance generalizability and

deepen understanding and explanation. This review utilized

the elements of the variable-oriented approach described by

Miles et al. (2014) through an examination of variables

within each study and their interrelationships. Research arti-

cles were read and critically reviewed several times.

Matrices were developed and analyzed for cross-case and

within-case analysis that included identifying levels of

evidence, study design, quality, sample sizes, sample char-

acteristics, measure of PA, and relevant themes.

Parental role modeling of PA included the factors of

parents’ enjoyment of PA, parents’ frequency of PA, fam-

ilies’ use of sports as recreation, and parents’ use of their

own behavior to encourage PA (Davison et al., 2003).

Studies examining parental role modeling of PA defined

it as parents’ participation, enjoyment, and recreation in

PA in front of the child. Parental modeling of behaviors

believed to increase children’s PA included parental par-

ticipation in PA alone or with their children and parents’

arranging of activities that included PA such as walking

together, visiting the swimming pool, family sports

events, and skating together (Davison & Campbell,

2005). Logistic support of PA incorporated parents enrol-

ling children in sports, attending child sporting events,

and assisting with transportation to PA events (Davison

et al., 2003). Monitoring PA included keeping track of

how much PA the child performed (Gubbels et al., 2011).

Restriction of sedentary behaviors included the parents

intentional efforts to limit the amount of time that the

child spent being sedentary.

Records identified through database searching

(n = 731) Sc

re en

in g

In cl

ud ed

E

lig ib

ili ty

Id

en ti

fi ca

ti on

 

Additional records identified through citations

(n = 1)

Records after duplicates removed (n =407)

Records screened (n =407)

Records excluded (n =319)

Full-text articles assessed for eligibility

(n =88)

Full-text articles excluded, no measure of parenting or children’s PA, protocols, reviews

(n = 50)

Studies included in qualitative synthesis

(n = 8)

Studies included in quantitative synthesis

(n = 30)

Figure 1. Flow diagram of study selection.

70 The Journal of School Nursing 34(1)

 

 

Results

Sample

Sample characteristics of the 38 manuscripts reviewed are

presented in Table 2. Descriptive correlational studies of the

topic were abundant, with 27 of the studies (71.1%) utilizing those methods to investigate potential association between

parenting practices and children’s PA. Eight (21.1%) of the studies were qualitative. One (2.6%) was a quasi-experimental design (Davison, Jurkowski, Li, Kranz, & Lawson, 2013), one

was a randomized controlled trial (Gerards et al., 2015), and

one was a pilot randomized controlled trial (O’Connor, Hil-

mers, Watson, Baranowski, & Giardino, 2011).

Demographic Characteristics

The children’s ages in the study ranged from infancy to 19

years old. The majority of the studies (n ¼ 19) focused on

school-aged children ages 6–12 years. Five studies focused

on preschool-aged children, and one study included infants

and preschoolers. Sample size of participants in the studies

ranged from 21 to 3,175. Studies conducted outside of the

United States were abundant. Studies conducted within the

United States focused predominantly on White families with

three studies focused on predominantly African American

samples and three on predominantly Hispanic samples (see

Table 2 for demographic characteristics of the samples).

Data Collection and Measurement

A variety of tools were used to measure the influence of

parenting practices on children’s PA levels. Nine of the 30

quantitative studies (30%) used the Activity-Related Parent- ing Practices Scale developed by Davison, Cutting, and

Burch (2003), developed specifically to measure parents’

PA-related parenting practices influencing girls’ PA. The

Activity-Related Parenting Practices Scale demonstrated

moderate to high Cronbach’s a (.61 to .86) in all studies. The Activity-Related Parenting Practices Scale was the most

frequently used tool to assess parenting practices across the

studies included in this review.

Studies not using the Activity-Related Parenting Prac-

tices Scale used a range of instruments to operationalize

parenting practices. Four studies used the Child Feeding

Questionnaire to measure parenting practices influencing

children’s activity (Arredondo et al., 2006; Gerards et al.,

2015; Gubbels et al., 2011; Remmers et al., 2014). Average

reported reliability of the Child Feeding Questionnaire was

0.67 (Remmers et al., 2014). Two of the quantitative studies

used the Parenting Strategies for Eating and Activities Scale

(PEAS). The PEAS included items about parental monitor-

ing of PA and parental reinforcement of PA. Reliability of

the PEAS for the included studies ranged from .7 to .88

(Lloyd, Lubans, Plotnikoff, Collins, & Morgan, 2014).

Ostybe and colleagues (2013) used the Home Environ-

ment Survey (HES) to measure parenting practices related to

PA in the physical, political, and sociocultural home envi-

ronment. The HES was the most detailed and inclusive

instrument used to measure parenting practices. Cronbach’s

a for items of the HES ranged between .65 and .8.

Quality

Eighteen of the 30 quantitative studies included in the

review were graded as low-quality evidence, as a result of

the majority of the studies were observational (see Table 1

for quality rankings of all quantitative studies in the review).

Accelerometers have been endorsed as the most reliable

and valid measure of PA in children (Eston, Rowlands, &

Ingledew, 1998). Fifteen of the 30 (50%) quantitative studies used accelerometers to measure children’s PA levels and 14

of the quantitative studies (46.7%%) used self-report scales as to measure PA. One (3.3%) of the quantitative studies used pedometers to measure PA (Lloyd et al., 2014).

Table 1. Quality of quantitative studies using GRADE criteria.

Studies GRADE Rating

Arredondo et al. (2006) Low Barnes, Plotnikoff, Collins, and Morgan (2015) Low Bradley, McRitchie, Houts, Nader, and O’Brien

(2011) Moderate

Crawford et al. (2010) Moderate Davison, Cutting, and Birch (2003) Low Davison et al. (2012) Low Davison, Jurkowski, Li, Kranz, and Lawson (2013) Moderate De Lepeleere, De Bourdeaudhuij, Cardon, and

Verloigne (2015) Low

Edwardson and Gorely (2010) Low Forthofer, Dowda, McIver, Barr-Anderson, and Pate

(2015) Moderate

Gerards et al. (2015) Moderate Gubbels et al. (2011) Low Heitzler, Martin, Duke, and Huhman (2006) Low Hennessy, Hughes, Goldberg, Hyatt, and

Economos (2010) Low

Jago et al. (2011) Moderate King et al. (2010) Low Lam and McHale (2014) Low Langer, Crain, Senso, Levy, and Sherwood (2014) Moderate Lloyd, Lubans, Plotnikoff, Collins, and Morgan (2014) Low Loprinzi, Schary, Beets, Leary, and Cardinal (2013) Low O’Connor, Chen, Baranowski, Thompson, and

Baranowski (2013) Low

O’Connor, Hilmers, Watson, Baranowski, and Giardino (2011)

Moderate

Ostbye et al. (2013) Low Remmers et al. (2014) Moderate Rhodes et al. (2015) Moderate Seabra et al. (2012) Low Tandon et al. (2014) Moderate Taylor, Wilson, Slater, and Mohr (2011) Low Verloigne, Van Lippevelde, Maes, Brug, and De

Bourdeaudhuij, (2012) Low

Zhao, Gao, and Settles (2013) Moderate

Hutchens and Lee 71

 

 

T a b

le 2 .

P ar

en ti n g

p ra

ct ic

es st

u d ie

s as

se ss

in g

p h ys

ic al

ac ti vi

ty o u tc

o m

es in

ch ild

re n .

A u th

o r,

Y ea

r

D es

ig n

Le ve

l o f

E vi

d en

ce Sa

m p le

P u rp

o se

P ar

en ti n g

P ra

ct ic

es V

ar ia

b le

s C

h ild

re n ’s

P h ys

ic al

A ct

iv it y

(P A

) V

ar ia

b le

s Si

gn ifi

ca n t

R es

u lt s

A rr

ed o n d o

et al

. (2

0 0 6 )

C ro

ss -s

ec ti o n al

, d es

cr ip

ti ve

co rr

el at

io n al

Le ve

l IV

n ¼

8 1 2

La ti n o

M ea

n ag

e ¼

6 ye

ar s

P ar

en ts

: 9 6 %

fe m

al e,

4 %

m al

e 4 8 %

B o ys

an d

5 2 %

gi rl

s 4 0 %

W it h

in co

m e

≤ $ 1 ,5

0 0

p er

m o n th

U n it ed

St at

es

E va

lu at

e th

e in

flu en

ce o f

p ar

en ti n g

st yl

e o n

ch ild

re n ’s

P A

an d

d ie

t

C h ild

Fe ed

in g

Q u es

ti o n n ai

re M

o n it o ri

n g

P A

: K

ee p in

g tr

ac k

o f am

o u n t

o f T

V w

at ch

ed an

d ex

er ci

se ch

ild is

ge tt

in g

P ar

en ta

l re

in fo

rc em

en t

o f P A

: P ra

is e

ch ild

fo r

b ei

n g

ac ti ve

D is

ci p lin

e th

e ch

ild fo

r sc

re en

ti m

e ac

ti vi

ti es

w it h o u t

p er

m is

si o n

Li m

it se

tt in

g: Li

m it

sc re

en ti m

e ac

ti vi

ti es

P ar

en t

co m

p le

te d

q u es

ti o n n ai

re P ar

en ta

l m

o n it o ri

n g

o f P A

w as

si gn

ifi ca

n tl y

as so

ci at

ed w

it h

ch ild

re n ’s

P A

P ar

en ta

l re

in fo

rc em

en t

o f P A

w as

si gn

ifi ca

n tl y

as so

ci at

ed w

it h

ch ild

re n ’s

P A

B ar

n es

, P lo

tn ik

o ff ,

C o lli

n s,

an d

M o rg

an (2

0 1 5 )

C ro

ss -s

ec ti o n al

, d es

cr ip

ti ve

co rr

el at

io n al

Le ve

l IV

n ¼

4 0 , m

o th

er s

an d

d au

gh te

rs M

ea n

ag e ¼

8 .8

ye ar

s P ar

en ts

: 1 0 0 %

fe m

al e

1 0 0 %

G ir

ls SE

S n o t

re p o rt

ed A

u st

ra lia

A ss

es s

m at

er n al

co rr

el at

es o f

o b je

ct iv

el y

m ea

su re

d P A

in gi

rl s

A ct

iv it y

Su p p o rt

(A C

T S)

Sc al

e Lo

gi st

ic su

p p o rt

M o d el

in g

G T

3 X

ac ce

le ro

m et

er s

Lo gi

st ic

su p p o rt

w as

p o si

ti ve

ly as

so ci

at ed

w it h

M V

P A

B en

tl ey

et al

. (2

0 1 2 )

Q u al

it at

iv e

st u d y

Le ve

l IV

n ¼

3 2 , p ar

en ts

6 –

to 8 -y

ea r-

o ld

s P ar

en ts

: 9 0 .6

% fe

m al

e, 9 .4

% m

al e

6 2 %

G ir

ls , 3 8 %

b o ys

SE S

U n it ed

K in

gd o m

Fo rm

at iv

e w

o rk

to in

fo rm

th e

co n te

n t

o f a

p ilo

t ra

n d o m

iz ed

co n tr

o l tr

ia l

(R C

T )

N A

N A

P ar

en ta

l lo

gi st

ic su

p p o rt

o f P A

fa ci

lit at

ed ch

ild ’s

P A

B ra

d le

y, M

cR it ch

ie ,

H o u ts

, N

ad er

, an

d O

’B ri

en (2

0 1 1 )

C ro

ss -s

ec ti o n al

, d es

cr ip

ti ve

co rr

el at

io n al

Le ve

l IV

n ¼

8 0 1

7 7 .3

% W

h it e

9 –

to 1 5 -y

ea r-

o ld

s P ar

en ts

: 1 0 0 %

fe m

al e

4 9 .4

% B o ys

, 5 0 .6

% gi

rl s

2 6 .1

% Lo

w SE

S U

n it ed

St at

es

E x am

in e

co n n ec

ti o n s

b et

w ee

n p ar

en ti n g

an d

o b je

ct iv

el y

m ea

su re

d P A

in 9 –

to 1 5 –

ye ar

-o ld

s

P ar

en ta

l en

co u ra

ge m

en t

o f

ch ild

P A

, p ar

ti ci

p at

in g

in P A

w it h

"Is this question part of your assignment? We can help"

ORDER NOW