Journal of Pediatric Nursing (2011) 26, 404–415

Use of a Tool to Determine Perceived Barriers to Children's

Healthy Eating and Physical Activity and Relationships to
Health Behaviors
Julie C. Jacobson Vann PhD, MS, RN a,⁎, Joanne Finkle JD b , Alice Ammerman DrPh c ,
Steven Wegner MD b,d , Asheley Cockrell Skinner PhD b ,
John T. Benjamin MD b , Eliana M. Perrin MD b
a
University of North Carolina at Chapel Hill, School of Nursing, Health Care Systems, Chapel Hill, NC
b
Division of General Pediatrics and Adolescent Medicine, Department of Pediatrics, The University of North Carolina at
Chapel Hill, School of Medicine, Chapel Hill, NC
c
The University of North Carolina at Chapel Hill, Center for Health Promotion Disease Prevention, Chapel Hill, NC
d
AccessCare, 3500 Gateway Centre Boulevard, Suite 130, Morrisville, NC

Key words:
This pilot investigation assesses whether barriers to children's healthy eating and physical activity
Obesity;
reported by parents on a newly developed brief pediatric obesity screening and counseling tool are
Healthy weight;
related to healthy eating and physical activity behaviors. The sample included parents of 115 Medicaid-
Barriers;
enrolled children in a general pediatric clinic. Of 10 barriers, 7 were statistically associated with parent-
Behaviors;
reported behaviors with odds ratios (ORs) ranging from 0.6 to 9.4. Relationships remained significant
Nutrition;
when child characteristics were controlled in the analysis. Although additional testing is needed, the
Physical activity
tool provides clinicians with an approach to identify barriers and behaviors for targeted counseling.
© 2011 Elsevier Inc. All rights reserved.

UNITED STATES' YOUTH are feeling the effects of an increased risk for becoming overweight or obese in
obesogenic society. An estimated 21.2% of 2- to 5-year-olds adulthood, with a myriad of associated health problems
and 35.5% of 6- to 11-year-olds are considered overweight (Baker, Olsen, & Sorensen, 2007; Dietz, 1998). For society,
or obese (Ogden, Carroll, Curtin, Lamb, & Flegal, 2010). the direct and indirect financial costs of childhood
This is approximately double and triple, respectively, the overweight and obesity add to the already overwhelming
prevalence observed 30 years ago in these age groups economic burden stemming from adult overweight and
(Koplan, Liverman, Kraak, & Committee on Prevention of obesity (Finkelstein, Trogdon, Cohen, & Dietz, 2009;
Obesity in Children and Youth, 2005). For children, excess Powers, Rehrig, & Jones, 2007).
weight increases the risk for early-onset chronic health Dietary behaviors, including consumption of sweetened
problems and potentially decreases social, psychological, beverages, excessive serving sizes, intake of fast food, and
and physical dimensions of quality of life (Schwimmer, insufficient consumption of fruits and vegetables, have been
Burwinkle, & Varni, 2003; Williams, Wake, Hesketh, associated with childhood overweight (Agras & Mascola,
Maher, & Waters, 2005). Overweight children are also at 2005; Dubois, Farmer, Girard, & Peterson, 2007; Kang et al.,
2006; Lioret, Volatier, Lafay, Tourvier, & Maire, 2009;
Taveras et al., 2005). Increases in “screen time” and a
⁎ Corresponding author: Julie C. Jacobson Vann, PhD, MS, RN. decrease in physical activity behaviors are also related to
E-mail address: jvann@email.unc.edu (J.C. Jacobson Vann). excess weight gain in children (Andersen, Crespo, Bartlett,

0882-5963/$ – see front matter © 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.pedn.2010.10.011
Barriers to Healthy Weight 405

Cheskin, & Pratt, 1998; Kang et al., 2006). Recently, known risks of childhood overweight and obesity. Barriers
these behaviors have often been attributed to a “toxic that are shown to be associated with physical activity and
environment” that encourages larger portion sizes, sugar- nutrition health behaviors may be important for clinicians
and fat-laden foods, parents' long workdays, and less to address with families in promoting healthy behaviors or
active lifestyles (Stettler, Signer, & Suter, 2004; Sturm, for communities to intervene at the population level.
2005a, 2005b).
In published studies, parents have identified barriers to
overcoming such a toxic environment when trying to feed Theoretical Framework
their children healthy foods or help their children to be more
physically active. Within the Health Belief Model, the The web of causation, Health Belief Model, Rogers'
concept of barriers has been defined as “one's opinion of the diffusion of innovations model, and transtheoretical model
tangible and psychological costs of the advised action” guided the development of the healthy eating and physical
(National Cancer Institute [NCI], 2005, p. 13). Several activity screening and counseling tool (Adams, 2009;
barriers to feeding children healthy food have been identified Centers for Disease Control and Prevention [CDC], 2007;
in published studies, including cost, preparation time, and Clement, 2008; De Civita & Dasgupta, 2007; Duncan, 2007;
children's food preferences (Ard et al., 2007; Hart, Herriot, Fahey & Burbridge, 2008; Mausner & Bahn, 1974, chap. 2;
Bishop, & Truby, 2003; Kicklighter et al., 2007; Omar, NCI, 2005; Stanhope, & Lancaster 1996, chap. 13; Wofford,
Coleman, & Hoerr, 2001). Cost and time have also been 2008). This tool is called Starting the Conversation (STC).
identified as barriers for parents trying to support children's The web of causation, originally conceived by MacMahon et
increased physical activity (Gordon-Larsen et al., 2004; al., essentially proposes that diseases or effects develop as
Irwin, He, Bouck, Tucker, & Pollett, 2005; McGarvey et al., the result of multiple factors or causes, each of which also
2006). Other barriers to physical activity include neighbor- results from “a complex genealogy of antecedents” (Duncan,
hood safety concerns and children's own fears related to 2007; Mausner & Bahn, 1974, chap. 2). The large number of
bullies on playgrounds (Gable, Chang, & Krull, 2007; antecedents creates the web. Individual-, family-, and
Gordon-Larsen et al., 2004; McGarvey et al., 2006; Weir, community-level factors are important to the development
Etelson, & Brand, 2006). of obesity (Wofford, 2008). This model, combined with
Pediatric nurses and primary care providers are well review of evidence from the literature and obesity experts,
positioned to identify and intervene with children who are has guided the selection of items for tool. The barriers in the
at risk for or are already overweight to increase healthful tool are expected to be antecedents to eating and physical
eating and physical activity behaviors. However, in today's activity behaviors.
fast-paced health services environment, it can be challeng- The Health Belief Model focuses on health behavior
ing for care providers to allocate their limited time to changes and incorporates four original and two newer
competing demands. Even when a clinician's intent to constructs: perceived susceptibility, perceived severity,
perform a behavior is high, obstacles may interfere with perceived benefits, perceived barriers, cues to action, and
carrying out the behavior (Perkins et al., 2007). Good self-efficacy (NCI, 2005). The screening and counseling
clinician intentions can be supported with systems that are tool was designed, in part, to identify perceived barriers
likely to facilitate performance improvement. For example, that may prevent people from taking positive action to
toolkits may be useful for improving provider involvement changing health behaviors (NCI, 2005). This is an
in counseling patients on health behaviors if the tools important first step in helping families develop plans of
address barriers to implementation of intended actions action that reduce the adverse effects of barriers. The STC
(Perkins et al., 2007). The North Carolina Medicaid was designed with “tips” for helping children eat well and
program sought to provide clinicians in the program with be physically active. These tips function as cues to action,
simple and effective time-saving tools that would combine providing parents with strategies to improve their children's
and match screening questions with counseling tips for health behaviors and remind them of negotiated plans for
parents. The tools were intended to assist clinicians with behavior changes (NCI, 2005).
assessing children's health behaviors and providing parents Rogers' diffusion of innovation model suggests that the
with targeted health education related to preventing and adoption of an innovation involves the interaction of the
treating overweight and obesity. A healthy eating and individuals adopting the innovation (the clinicians) and the
physical activity screening and counseling tool was innovation itself, the STC (De Civita & Dasgupta, 2007).
developed as a component of the healthy weight toolkit. Within this model, the rate of adoption of the innovation is
Development of this tool is described below. The objective dependent on the characteristics of the innovation, the
of this secondary data analysis of a prior intervention was methods used to communicate, the characteristics of the
to conduct preliminary testing of the newly developed rapid users, and the social systems in which the innovation is
assessment and counseling tool. The testing involved introduced (Adams, 2009). The STC was designed to address
assessing whether the barriers to healthy eating and the concepts in the model of relative advantage, compatibil-
physical activity are associated with behaviors that reflect ity, and complexity (Fahey & Burbridge, 2008) by creating a
406 J.C. Jacobson Vann et al.

tool that can be efficiently applied in a fast-paced providers (Perrin, Finkle, & Benjamin, 2007). The STC-4-12
environment, combining screening with counseling strate- tool was designed with the goal of facilitating efficient,
gies, and breaking down the components into manageable consistent, and structured yet individualized assessment and
parts. The characteristics of the tool are intended to facilitate counseling by clinicians in a fast-paced environment. The
adoption of the STC tool by clinicians. Nutrition STC-4-12 includes five “barrier” questions, five
The transtheoretical model specifies the stages of change nutrition behavior questions, a parental “readiness to change”
or degree of readiness to change or begin new behaviors question, and two to three “tips to help your child eat well” for
(Clement, 2008). The stages in the model include precon- each behavior and barriers nutrition question. Barrier questions
templation, contemplation, preparation, action, and mainte- were intended to measure beliefs and behaviors that may
nance (CDC, 2007; Henry, 2005). The STC also included inhibit parents from helping their children engage in behaviors
nutrition and physical activity readiness to change questions that are likely to help promote healthy weight. The Physical
to assist clinicians with identifying the parents' receptiveness Activity STC-4-12 consists of five physical activity “barriers”
to helping their children improve health behavior and questions, five physical activity behaviors questions, one
intervene based on readiness. parental readiness to change question, and two to three “tips to
help your child be more active” for each physical activity
Methods question. The tips for helping children eat well and be more
active are matched with each behavior and barrier question to
guide counseling by clinicians and serve as take-home
Study Design and Participants
reminders for parents. All except one of the STC-4-12 barriers
and behaviors questions included response scales with three
Pediatric patients and their parents were recruited
choices. The question asking about type of milk consumed
between April 2005 and March 2006 at the University of
most often added a fourth response option: “none.” The STC-
North Carolina at Chapel Hill (UNC-CH), Child and
4-12 questions and response options are listed in Appendix 1.
Adolescent General Clinic for an exploratory correlational
The STC-4-12 tool was developed through a yearlong
(Brink & Wood, 1997) pilot study of a brief intervention iterative qualitative consensus process by a team of nurses,
aimed at preventing and treating childhood obesity in a
pediatricians, nutritionists, and epidemiologists with exper-
primary care setting. Parents and their children were
tise in childhood obesity. Prior to developing the STC-4-12,
eligible to participate if the children were 4 to 12 years
currently available nutrition and physical activity education,
of age, seen in the clinic for a well-child visit or minor
assessment, and counseling tools developed for clinician use
illness, and insured by North Carolina Medicaid or the
were obtained from Internet searches, clinical practices
State Children's Health Insurance Program (SCHIP). The
throughout North Carolina, and out-of-state clinics known
study was restricted to English-speaking parents because
for obesity assessment and counseling measures. Focus
this was a pilot study and the tool of interest had not yet groups were held with clinicians at six primary care practices
been translated for families speaking other languages. This
to explore experiences with existing obesity screening and
study was approved by the UNC-CH School of Medicine
counseling tools, perceived strengths and weaknesses of each
Institutional Review Board (Protocol 04-HPDP-771).
tool, and attributes of tools that would be helpful for them to
assess and counsel overweight and obese patients. Focus
Study Procedures group responses were summarized qualitatively by theme.
Results of focus groups were published previously (Flower,
Clinic appointment schedules were reviewed to identify Perrin, Viadro, & Ammerman, 2007). Extensive literature
potentially eligible children. All potentially eligible parents searches were conducted on factors associated with healthy
and children were approached in private examination rooms eating, physical activity, and childhood overweight and
by a research associate. Study objectives and procedures obesity. Identified instruments, evidence of instrument
were explained to families; informed consent was obtained effectiveness, relevant literature, and focus group responses
from parents. Assent was obtained from children at least were reviewed and qualitatively analyzed by the develop-
7 years of age. Parents were asked to complete the two-page ment team. It was determined that few existing tools were
STC tool (described under Study Instruments and Measure- culturally sensitive for minority and low-income populations
ment) and a brief intake tool consisting of contact and and of the appropriate literacy level for Medicaid populations
demographic information. or fostered interactive exchange between the patient, parent,
and provider. In addition, many tools did not appear to be
Study Instruments and Measurement based on evidence or theory.
The format of STC-4-12 was based on previously
The Starting the Conversation Nutrition and Physical developed STC tools for adults (STC-adult) that focused
Activity tool for 4- to 12-year-olds (STC-4-12) is a two-part, on healthy eating and physical activity (NC Prevention
evidence-informed rapid assessment tool and tailored counsel- Partners, 2010; Gaskins et al., 2007; Glasgow et al., 2005).
ing guide designed for use by nurses and other primary care Approximately 60% of the STC-4-12 questions were
Barriers to Healthy Weight 407

obtained from the adult healthy eating and physical activity for age, gender, race, and ethnicity of the children. ORs can
STCs, with questions modified for children as necessary. The be used to describe the strength of associations. An OR of
STC-adult development involved extensive formative and 1.0 indicates no observed statistical association between the
pilot testing. The adult dietary STC was validated using barrier and behavior. An OR greater than 1.0 indicates a
comparisons to the longer validated 54-item Dietary Risk positive relationship, and an OR less than 1.0 indicates a
Assessment (DRA) Scores and serum carotenoid levels negative relationship. The OR is difficult to interpret directly.
(Paxton, Ammerman, Gizlice, Johnston, & Keyserling, However, because the OR is often considered an estimate of
2007). The DRA score and total STC score were highly the relative risk, it is easiest to interpret the OR as a relative
correlated (r = .67, p b .0001). The STC-4-12 was pretested risk or risk ratio for simplicity. A simplistic interpretation of
with parents and clinicians at two primary care clinics during an OR of 9.4 in this study is as follows: parents who agree that
a 3-week period by a nutritionist and nurse investigator. their children do not like healthy foods are 9.4 times more
Feedback from clinicians and parents was used to modify likely to report that their children eat two or fewer servings of
and increase the clarity of questions and response options. fruits and vegetables compared with parents who disagree
Internal consistency of the STC-4-12 was assessed through with the statement. A CI for an OR that does not include 1.0
computation of Cronbach's alpha with an overall scale is considered a statistically significant result. Because the
reliability coefficient of .75, calculated using all items in their variables are somewhat dependent on one another, a
original scales (Carmines & Zeller, 1979, chap. 4). Bonferroni's correction was not applied, as the corrections
would be expected to be highly conservative and miss real
associations, increasing type II errors (Garamszegi, 2006).
Data Management and Analysis

Teleform software was used to facilitate data collection Results

and entry and export of the parent survey and intake tool. The
STC forms were printed as three-part carbonless forms. Study Sample and Patient and
These data were manually entered into a database. Statistical Parent Characteristics
analyses were conducted using Stata 10 (College Park, TX)
and SAS 9.1.3 (Cary, NC). The mean age of participating children was 7.5 years at
enrollment (Table 1). Of the 115 enrolled children,
Statistical Analysis approximately half were female. Most children were non-
Hispanic (92%), self-identified as Black or African Amer-
ican (66%), and insured by Medicaid (82.6%). The children
The gamma statistic was used to assess the strength and
and their parents primarily lived in urban areas, based on
direction of relationships between pairs of STC barrier
and behavior measures, with statistical significance deter- Rural–Urban Commuting Area codes (Hart, Larson, &
Lishner, 2005). More than one third of the children (34.8%)
mined by 95% confidence intervals (CIs). The gamma
were obese (body mass index [BMI] ≥ 95th percentile), and
statistic or coefficient is a correlation measure, equivalent
14.8% were overweight (BMI ≥ 85%; Barlow & Expert
to Spearman's R or Kendall tau. This measure is appro-
Committee, 2007). Most (92.2%) of the responding parents
priate for ordinal categorical data and is preferable when
were female, and more than one half (54.8%) of the parents
the data contain many tied observations. Gamma takes
considered themselves to be overweight (Table 1).
values between −1 and 1. A gamma value less than zero
indicates a negative association; a gamma value greater
than zero indicates a positive association; and a gamma Barriers to Healthy Eating
value of zero means that the measures are independent
(Stevens et al., 2004). Approximately half of the parents agreed “a little” or
The potential influence of demographic characteristics on “a lot” to the statements “My child likes to eat in front of
relationships between barriers and behaviors was assessed the TV or at the computer” and “Healthy foods cost
using logistic regression. For ease of interpretation, some of too much” (Table 2). In addition, 36.5% of the parents
the response categories for dependent (behavior) and agreed “a little” or “a lot” with the statement “My child
explanatory (barrier) variables were combined to create doesn't like healthy foods.”
two response categories instead of three. For example, the
barriers response categories agreeing “a little” and “a lot” Barriers to Physical Activity
were combined to compare to those who disagreed at all with
barrier statements. Unadjusted ORs were computed to repre- More than one half of the parents agreed “a little” or “a
sent the effect of agreeing “a little” or “a lot” to a barrier on lot” to the statements “When my child misbehaves, I take
each behavior. Adjusted ORs (aORs) and CIs were computed away their outdoor/indoor play time” (55.7%) and “It's hard
using the results of logistic regression to assess the rela- for me to find time to play outside with my child” (50.4%;
tionships between barriers and behaviors while controlling Table 2). In addition, more than 40% of parents agreed with
408 J.C. Jacobson Vann et al.

Table 1 Characteristics of Children and Their Parents behavior, based on a modest to strong gamma statistic.
n = 115, The parental belief that “Sometimes it seems like the only
n (%) M (SD) way to get my child to behave is to promise candy or other
food treats” was associated with reports that children ate
Child's age (years) 7.5 (2.8)
fewer servings of fruits and vegetables (γ = 0.50, 95% CI =
Child's sex, male 58 (50.4)
0.18–0.82) and ate food purchased away from home more
Child's ethnicity
Hispanic/Latino 9 (7.8) often (γ = 0.33, 95% CI = 0.16–0.65). The parental
Not Hispanic/Latino 106 (9.2) perception that “My child doesn't like healthy foods” was
Child's race strongly associated with children eating less servings of
American Indian/Alaskan Native 2 (1.7) fruits and vegetables (γ = 0.75, 95% CI = 0.58–0.92) and
Asian 1 (0.9) moderately associated with eating a greater number of
African American 76 (66.1) junk food snacks (γ = 0.43, 95% CI = 0.18, 0.68). Logistic
White 24 (20.9) regression modeling supported the significant gamma
Other 12 (10.4) correlation results (Figure 1) and demonstrated that
Child's insurance coverage significant associations observed between healthy eating
Health choice (SCHIP) 20 (17.4)
barriers and behaviors remained generally consistent after
Medicaid 95 (82.6)
adjusting for patient characteristics. The strongest asso-
Body mass index classification of child
Healthy weight (5th to b85th percentile) 58 (50.4) ciation was observed between the barrier “My child doesn't
Overweight (85 to b95th percentile) 17 (14.8) like healthy foods” and lower reported fruit and vegetable
Obese (≥95th percentile) 40 (34.8) consumption (aOR = 9.4, 95% CI = 3.5–24.9). Parental
Gender of parent, female 106 (92.2) agreement with the statement “Sometimes it seems like
Parent's classification of own weight the only way to get my child to behave is to promise candy
Underweight 1 (0.9) or other food treats” was significantly related to three of
Healthy weight 51 (44.4) four tested behaviors, with adjusted ORs ranging from 3.1
Overweight 63 (54.8) to 7.6 (Figure 1).

the statement “It takes too much time and money to have Relationships Between Barriers to Physical Activity
my child involved with sports programs.”
and Physical Activity-Related Behaviors
Relationships Between Barriers to Healthy Eating Four of five queried barriers to physical activity were
and Nutrition-Related Behaviors significantly associated with at least one physical activity-
related behavior. Parents who reported that “It's hard for me
Three of five queried dietary barriers were found to be to find time to play outside with my child” were more likely
significantly associated with at least one healthy eating to report fewer “hours of active play” by children (γ = 0.57,

Table 2 Distribution of Potential Barriers to Healthy Eating and Physical Activity (n = 115)
Agree
“A Little”/ Agree Agree
Disagree “A Lot” “A Little” “A Lot”
Variable n (%) n (%) n (%) n (%)
Healthy eating barriers
Promise candy or other food treats to get child to behave 90 (78.3) 25 (21.7) 15 (13.0) 10 (8.7)
Healthy food costs too much 67 (58.3) 48 (41.7) 28 (24.3) 20 (17.4)
Child doesn't like healthy food 73 (63.5) 42 (36.5) 35 (30.4) 7 (6.1)
Healthy meals take too long to prepare 102 (88.7) 13 (11.3) 10 (8.7) 3 (2.6)
Child likes to eat in front of television or at the computer 58 (50.4) 57 (49.5) 32 (27.8) 25 (21.7)
Physical activity barriers
Take away outdoor/indoor play time when child misbehaves 51 (44.3) 64 (55.7) 43 (37.4) 21 (18.3)
Hard to find time to play outside with child 56 (49.1) 58 (50.8) 38 (33.3) 20 (17.5)
Takes too much time and money to have child involved 67 (58.3) 48 (41.7) 36 (31.3) 12 (10.4)
with sports programs
Child feels that s/he will get teased when playing outside 100 (87.0) 15 (13.1) 8 (7.0) 7 (6.1)
or on a team
Too busy to drive child to activities and sports 97 (84.3) 18 (15.6) 12 (10.4) 6 (5.2)
Barriers to Healthy Weight 409

Figure 1 Adjusted ORs of the effect of agreeing “a little” or “a lot” to diet barriers on behaviors. An OR of 1.0 (the vertical line) indicates
no association between the barrier and behavior. A CI that does not cross 1.0 would indicate a statistically significant OR.

95% CI = 0.35–0.79), fewer days in which the “family or involvement in “school sports teams or community groups”
community do active things together” (γ = 0.44, 95% CI = (γ = 0.33, 95% CI = 0.06–0.59) than parents who disagreed
0.20–0.68), more hours of child's “screen time” (γ = 0.39, with the barrier statement about finding time to play with
95% CI = 0.17–0.62), fewer days playing outdoors (γ = their children. The belief that “My child feels s/he will get
0.39, 95% CI = 0.15–0.63), and less frequent child teased when playing outside or on a team” was associated
410 J.C. Jacobson Vann et al.

with fewer days of playing outdoors per week (γ = 0.49, preferences was modified by the availability of fruits and
95% CI = 0.11–0.86) and fewer days per week in which vegetables (Neumark-Sztainer et al., 2003). Even when fruit
the “family or community do active things together” (γ = and vegetable taste preferences were reported to be low,
0.40, 95% CI = 0.14–0.68). The concern that “It takes too self-reported intake increased, on average, when fruits and
much time and money to have my child involved with vegetables were made available (Neumark-Sztainer et al.,
sports programs” was related to less frequent participation 2003), for example, through having them in view of the
in active family or community events (γ = 0.41, 95% CI = adolescent (Story, Neumark-Sztainer, & French, 2002). A
0.14–0.68). Findings from logistic regression modeling focus group-based study of 213 children with ages 7 to 17
support the significant correlation analyses (γ) for the years also ranked internal/physiologic preferences such as
proposed barriers: “It's hard for me to find time to play taste preferences and cravings to be a barrier to healthful
outside with my child”; “It takes too much time and money eating (O'Dea, 2003). Similar relationships were identified
to have my child involved with sports programs”; and in a published review of the literature (Jenkins & Horner,
“When my child misbehaves, I take away their outdoor/ 2005) and other studies of children and adolescents (Hart
indoor play time.” Parental agreement with the statement et al., 2003; Kicklighter et al., 2007; Neumark-Sztainer,
“It's hard for me to find time to play outside with my Story, Perry, & Casey, 1999).
child” was associated with reports of lower levels of acti- In this study, 41.7% of parents agreed “a little” or “a lot”
vity for all five physical activity behaviors, with adjusted to the statement “Healthy foods cost too much.” However,
ORs ranging from 2.2 to 5.5 (Figure 2). this perceived barrier to healthy eating was not associated
significantly with eating behaviors included in the STC-4-12.
This finding was unexpected because cost has been identified
as a barrier to healthy eating in other published studies. In a
Discussion study of households of 1,355 children in Alabama, higher
costs of fruits and vegetables, as measured by the
In this study, statistically significant relationships were Agricultural Research Service of the United States Depart-
observed between 7 of 10 proposed barriers to healthy eating ment of Agriculture, was inversely related to availability in
and physical activity and one or more health behaviors the homes (Ard et al., 2007). Other studies employing focus
included in the brief STC-4-12 screening and counseling groups of parents or grandparents caring for children cited
tool. Observed statistical relationships were particularly cost as a barrier to healthy eating (Hart et al., 2003;
strong for the barriers that children do not like healthy Kicklighter et al., 2007; Omar et al., 2001; Story et al., 2002);
foods and parents have difficulty finding time to play outside yet, one study reported that parents may have had
with their children. Parents who reported that their children misperceptions about what are considered to be healthy
did not like healthy foods were more likely to report that their foods (Hart et al., 2003). It is possible that the lack of
children consumed fewer servings of fruits and vegetables statistical association between perceived cost and healthy
and ate two or more unhealthy snacks per day. Parents who eating in this study is related to parental misperceptions
agreed with the statement that it is “hard to find time to about healthy foods or that parents may strive to overcome
play outside with my child” were more likely to report that perceived barriers such as cost.
their children had fewer hours of active play, more screen The general lack of statistical association in this study
time, rare or no involvement in sports or community between the perception that healthy foods take too long
groups, and infrequent participation in active community or to prepare and healthy eating behaviors was unexpected
family events. Other perceived barriers that show signifi- and inconsistent with studies that examined time as a
cant statistical relationships with unhealthy behaviors potential barrier to healthy eating. In a study of male and
include parents' beliefs that they need to promise food female caregivers of young children in three rural counties
treats to get children to behave, parents' beliefs that their in Michigan, scarcity of time was specified as one of three
children like to eat in front of the television or computer, major barriers to providing healthy meals (Omar et al.,
parental practice of taking away play time if children 2001). Caregivers cited demands of outside work, as well
misbehave, and concerns that it takes too much time and as time to plan, shop, and prepare healthy meals (Omar
money for sports programs. et al., 2001). In this study, most parents (88.7%) disagreed
In this study, the significant relationship observed with the statement that healthy meals take too long to
between parents' perceptions that their children do not like prepare. One possible explanation for the small proportion
healthy foods and reports of children consuming fewer of participants who reported that healthy meals take
servings of fruits and vegetables is supported by published too long to prepare may be the focus on healthy food
results of surveys and focus groups. In one survey of 4,746 preparation in the media through, for example, cooking
adolescents in Minnesota, one of the strongest correlates of programs. The relative lack of variability in the responses
fruit and vegetable consumption was taste preferences to this barriers question may help to explain the absence
(Neumark-Sztainer, Wall, Perry, & Story, 2003). However, of a statistically significant association between time as a
the relationship between fruit and vegetable intake and taste barrier and any of the healthy eating behavior questions.
Barriers to Healthy Weight 411

Figure 2 Adjusted ORs of the effect of agreeing “a little” or “a lot” to physical activity barriers on physical activity behaviors. An OR of
1.0 (the vertical line) indicates no association between the barrier and behavior. A CI that does not cross 1.0 would indicate a statistically
significant OR.

Time and/or money are perceived as barriers to children's parents in this study reported that it is hard to find time to
physical activity in published scientific literature (Gordon- play outside with their children, and 41.4% indicated that it
Larsen et al., 2004; Irwin et al., 2005; McGarvey et al., takes too much time and money to have children involved in
2006) and in this study. Approximately half (50.4%) of sports programs. These time and money barriers in the STC-
412 J.C. Jacobson Vann et al.

4-12 were significantly statistically associated with physical Limitations

activity behaviors, most consistently for the former time
barrier. In a North Carolina church-based pilot project of There are several important limitations to this study.
African American girls, with ages 6 to 9 years, and their First, the study sample was selected from only one clinic and
female caregivers, perceived environmental barriers, such was limited to those who spoke English and were enrolled in
as lack of affordable and accessible recreational resources, the State Medicaid or SCHIP, raising questions of general-
were believed to reduce opportunities for physical activity izability. For example, the item “Healthy foods cost too
(Gordon-Larsen et al., 2004). In a Canadian study of much” is likely to be less relevant to families of higher
parents of preschool children, time and finances were two socioeconomic status. Yet, generally, the behaviors and
of nine barriers and facilitators of adequate physical activity barriers constructs were identified from literature and
(Irwin et al., 2005). clinician experiences that included families of varying
The findings from this pilot study suggest that several socioeconomic backgrounds. Testing this screening and
parental beliefs and other potential barriers to healthy eating counseling tool in other populations, including those with
and physical activity play essential roles in the multifactorial private health insurance, is an important next step. In
web of causation related to the development of childhood addition, modified versions of the tool will need to be
overweight and obesity. A parent's belief related to healthy developed and tested for families with culturally diverse
food or physical activity, regardless of whether the belief backgrounds that may influence health behaviors and weight
matches reality, can either encourage or discourage healthy status through unique pathways.
behaviors. When a belief discourages a healthy behavior, the A second limitation of the study is the small sample size
belief becomes an internal barrier. Examples of internal and large CIs. A larger sample size may have increased the
barriers include beliefs such as using food as a reward for precision and therefore narrowed the CIs for study measures
good behavior will result in a more obedient child, preparing and increased the number of observed positive relationships
healthy foods takes too much time, or being physically active that were found to be statistically significant. To maximize
costs too much. The perception that a neighborhood is unsafe the sample size and minimize the CIs for study measures, all
for outdoor play can also be an internal barrier when the fear potentially eligible parents and their children were asked to
is not based in reality (Weir et al., 2006). Although some participate. In addition, parents rarely refused to participate.
healthy foods are expensive, a parent's belief that all healthy The small sample size may be attributed to the relatively
foods are too expensive may keep the parent from making infrequent occurrence of parents scheduling and bringing
dietary improvements. Family beliefs are important for their 4- to 12- year-olds to clinic for well-child visits in this
health care providers to consider when working with families population. It is also possible that the responses of parents
on healthy lifestyle changes. who take their 4- to 12-year-old children to well-child visits
Additional studies are needed to more clearly elucidate may differ from those who do not routinely take their child to
the relationships between perceived barriers to healthy eating well-child visits.
and physical activity, health behaviors, and weight status of A third limitation to be considered when interpreting
children. The results of this study add to a growing body of results is that this is a small preliminary pilot project using
research that seeks to further understand barriers to healthy an instrument that has not been fully statistically tested for
weight and identify, develop, and/or evaluate lifestyle reliability and validity, especially for the 40% of the items
interventions aimed at preventing and reducing childhood that were not directly taken from the adult STC. Internal
overweight and obesity (American Academy of Pediatrics consistency of the items in the STC-4-12 was tested
[AAP], 2003; Nemet et al., 2005; Wilfley et al., 2007). The through computation of Cronbach's alpha. Face validity
potential value of the STC-4-12 in promoting healthy weight was assessed by nurses, physicians, and nutritionists who
in children can be strengthened with further research and have extensive experience with childhood obesity. Pretesting
testing of the tool. The STC-4-12 should be tested with a was performed with both parents and primary care clini-
larger sample to continue to assess relationships between cians. Focus groups and published literature were used to
barriers and behaviors and to determine whether these identify and verify salient barriers and behaviors for rapid
relationships vary based on children's age. Additional testing screening, supporting content validity. Criterion-related
of the STC-4-12 should include comparisons of the validity was tested, in part, through statistically assessing
relationships between barriers and behaviors for children the relationships between the barriers and behaviors
who are overweight as compared with those who are not (Carmines & Zeller, 1979, chap. 2). However, additional
overweight. Longitudinal studies should be conducted to testing of the STC-4-12 is needed.
assess the potential effectiveness of clinician's use of the The list of barriers in the STC-4-12 was derived from
STC-4-12 tool with helping families improve targeted preliminary exploratory work through an iterative process
health behaviors and weight status of children. Until involving a team of pediatric health care professionals
additional testing is completed, this tool may be valuable working with community and clinic populations. Because
for clinicians to use to help initiate conversations with this tool was developed to assist clinicians with rapid
patients and families about health behaviors and barriers. assessment when time is significantly limited, the screening
Barriers to Healthy Weight 413

and counseling tool did not include a comprehensive associated with parental reports of their children's behaviors.
compilation of barriers. Although the intent was to include Several of the identified associations were supported by
only a limited number of items when designing the tool, it is previously published literature concerning barriers to healthy
possible that other barriers more strongly associated with eating and physical activity. Additional research is needed to
eating and physical activity may have been omitted. Ongoing assess whether the STC-4-12 screening and counseling tool
efforts should be made to identify critical potential barriers to is helpful for families in their efforts to improve health
healthy eating and physical activity. behaviors and prevent or reverse any resultant unhealthy
In addition, there is a potential for response bias, as weight trajectories. Recognizing the need for further testing,
parents may be more likely to underreport both barriers clinicians may find the tool useful for beginning the
and children's unhealthy behaviors to their children's conversations with parents and children about reducing
health care providers. If parental underreporting of barriers and increasing healthy behaviors.
barriers is systematically different than underreporting of
unhealthy behaviors, then the observed associations
between barriers and behaviors are likely to be biased. In Acknowledgments
addition, it was not determined if the magnitude of the
reported barriers were related to specific health behaviors. This project was supported by Dr. Perrin's NIH K23 career
Finally, it was not determined whether parents who development award (1K23 HD051817-01A1), Dr. Ammer-
perceived specific barriers had worked to overcome some man's R01 (5 R01 HD050981-02), and Dr. Skinner's
of the barriers. Postdoctoral Fellowship (5 T32 NR008856).
The authors would like to thank Lung-Chang Chien for
his assistance with graphics, Moonsu Kang for assistance
Conclusions with data analysis, Kori Flower, MD, MPH and Sari Teplin,
MS, MPH for their contributions to original tool develop-
This exploratory pilot study suggests that many of the ment, and Karah Daniels, MA, Gina Chung, EdM, and Lisa
proposed barriers to healthy eating and physical activity Pullen-Davis, PhD, MSPH for project planning and
questions included in the study instrument were strongly implementation support.

Appendix 1. Starting the Conversation Nutrition and Physical Activity, Barriers, and Behaviors
Questions, Ages 4–12
Question Response Options

Healthy eating behaviors

How many servings of fruits and/or vegetables does your child eat each day? 5 or more, 3–4, less than 3
How many sugary drinks (soda, sports drinks, sweet tea, lemonade, “Kool-Aid”) and fruit 1 or fewer, 2, more than 2
juice does your child drink each day?
How many snacks like cakes, cookies, ice cream, candy, and chips does your child eat each day? 1 or fewer, 2–3, more than 3
How many times a week does your child eat food purchased away from home (fast food, 1 or fewer, 2–3, more than 3
convenience stores, vending machines)?
The milk that my child most often drinks is… Skim or 1% milk, 2%, whole, none
Healthy eating barriers
Sometimes it seems like the only way to get my child to behave is to promise candy or other Disagree, agree a little, agree a lot
food treats.
“Healthy foods cost too much.” Disagree, agree a little, agree a lot
“My child doesn’t like healthy foods.” Disagree, agree a little, agree a lot
“I find that healthy meals take too long to prepare.” Disagree, agree a little, agree a lot
“My child likes to eat in front of the TV or at the computer.” Disagree, agree a little, agree a lot
Readiness to change (nutrition)
How do you feel about making some changes to help your child eat healthier? I am not interested in making
changes at this time.
I am not ready to make changes yet,
but want to talk more.
I am ready to make some changes
now and would like help.
I am already helping my child to eat
healthier and don't feel there is
much more to do.
414 J.C. Jacobson Vann et al.

Appendix 1 (continued)
Question Response Options
Physical activity behaviors
How many hours of active play does your child get every day? More than 2, 1–2, fewer than 1
How many days a week does your child play outdoors? 5 or more, 3–4, 0–2
How many hours of “screen time” (TV, video, computer games) does your child get 0–2, 3, more than 3
each day?
How often is your child involved in school sports teams or community groups like More than once a week, once a
basketball, swimming or step/dance? week, rarely/never
How many days a week does your family or community do active things together? More than 3, 2–3, less than 2
Physical activity barriers
“When my child misbehaves, I take away their outdoor/indoor play time.” Disagree, agree a little, agree a lot
“It's hard for me to find time to play outside with my child.” Disagree, agree a little, agree a lot
“It takes too much time and money to have my child involved with sports programs.” Disagree, agree a little, agree a lot
“My child feels that s/he will get teased when playing outside or on a team.” Disagree, agree a little, agree a lot
“I feel like I’m too busy to drive my child to activities and sports.” Disagree, agree a little, agree a lot
Readiness to change physical activity
How do you feel about making some changes to help your child be more active? I am not interested in making
changes at this time.
I am not ready to make changes yet,
but want to talk more.
I am ready to make some changes
now and would like help.
I am already helping my child to be
more active and don't feel there is
much more to do.

References change interventions. Retrieved July 1, 2009, from: http://www.cdc.

gov/std/Program/community/9-PGcommunity.htm#diffusion.
Clement, D. (2008). The transtheoretical model: An exploratory look at its
Adams, S. (2009). Use of evidence-based practice in school nursing: Survey applicability to injury rehabilitation. Journal of Sports Rehabilitation,
of school nurses at a national conference. The Journal of School 17, 269−282.
Nursing, 25, 302−313. De Civita, M., & Dasgupta, K. (2007). Using diffusion of innovations theory
Agras, W. S., & Mascola, A. J (2005). Risk factors for childhood to guide diabetes management program development: An illustrative
overweight. Current Opinions in Pediatrics, 7, 648−652. example. Journal of Public Health, 29, 263−268.
American Academy of Pediatrics. (2003). Committee on Nutrition. Policy Dietz, W. H. (1998). Health consequences of obesity in youth: Childhood
statement: Prevention of pediatric overweight and obesity. Pediatrics, predictors of adult disease. Pediatrics, 101(3 Pt 2), 518−525.
112, 424−430. Dubois, L., Farmer, A., Girard, M., & Peterson, K. (2007). Regular sugar-
Andersen, R. E., Crespo, C. J., Bartlett, S. J., Cheskin, L. J., & Pratt, M. sweetened beverage consumption between meals increases risk of
(1998). Relationship of physical activity and television watching with overweight among preschool-aged children. Journal of American
body weight and level of fatness among children: Results from the Dietetic Association, 107, 924−934.
third national health and nutrition examination survey. Journal of the Duncan, D. F. (2007). Epidemiology: Basis for disease prevention and
American Medical Association, 279, 938−942. health promotion. Multicausality and webs of causation. Retrieved
Ard, J. D., Fitzpatrick, S., Desmond, R. A., Sutton, B. S., Pisu, M., January 30, 2010, from: http://duncansepidemiology.tripod.com/id9.html.
Allison, D. B., et al. (2007). The impact of cost on the availability Fahey, D. F., & Burbridge, G. (2008). Application of diffusion of
of fruits and vegetables in the homes of schoolchildren in innovations models in hospital knowledge management systems:
Birmingham, Alabama. American Journal of Public Health, 97, Lessons to be learned in complex organizations. Hospital Topics, 86,
367−372. 21−31.
Baker, J. L., Olsen, L. W., & Sorensen, T. I. (2007). Childhood body-mass Finkelstein, E. A., Trogdon, J. G., Cohen, J. W., & Dietz, W. (2009). Annual
index and the risk of coronary heart disease in adulthood. New England medical spending attributable to obesity: Payer- and service-specific
Journal of Medicine, 357, 2329−2337. estimates. Health Affairs, 28, w822−w831.
Barlow, S. E., & Expert Committee. (2007). Expert committee recommen- Flower, K. B., Perrin, E. M., Viadro, C. I., & Ammerman, A. S. (2007).
dations regarding the prevention, assessment, and treatment of child Using body mass index to identify overweight children: Barriers and
and adolescent overweight and obesity: Summary report. Pediatrics, facilitators in primary care. Ambulatory Pediatrics, 7, 38−44.
120(Suppl 4), S164−S192. Gable, S., Chang, Y., & Krull, J. L. (2007). Television watching and
Brink, P. J., & Wood, M. J. (1997). Advanced design in nursing research. frequency of family meals are predictive of overweight onset and
Thousand Oaks, CA: Sage Publications. persistence in a national sample of school-aged children. Journal of
Carmines, E. G., & Zeller, R. A. (1979). Reliability and validity assessment. American Dietetics Association, 107, 53−61.
Newbury Park: Sage Publications. Garamszegi, L. Z. (2006). Comparing effect sizes across variables:
Centers for Disease Control and Prevention. (2007). Program operations Generalization without the need for Bonferroni correction. Behavioral
guidelines for STD Prevention, community and individual behavior Ecology, 17, 682−687.
Barriers to Healthy Weight 415

Gaskins, N. D., Sloane, P. D., Mitchell, C. M., Ammerman, A, Ickes, S. B., Neumark-Sztainer, D., Wall, M., Perry, C., & Story, M. (2003). Correlates
& Williams, C. S. (2007). Poor nutritional habits: A modifiable of fruit and vegetable intake among adolescents: Findings from Project
predecessor of chronic illness? A North Carolina Family Medicine EAT. Preventive Medicine, 37, 198−208.
Research Network (NC-FM-RN) study. Journal of the American Board O'Dea, J. A. (2003). Why do kids eat healthful food? Perceived benefits
of Family Medicine, 20, 124−134. of and barriers to healthful eating and physical activity among children
Glasgow, R. E., Ory, M. G., Klesges, L. M., Cifuentes, M., Fernald, D. H., & adolescents. Journal of the American Dietetic Association, 103,
& Green, L. A. (2005). Practical and relevant self-report measures of 497−501.
patient health behaviors for primary care research. Annals of Family Ogden, C. L., Carroll, M. D., Curtin, L. R., Lamb, M. M., & Flegal, K. M.
Medicine, 3, 73−81. (2010). Prevalence of high body mass index in US children and
Gordon-Larsen, P., Griffiths, P., Bentley, M. E., Ward, D. S., Kelsey, K., adolescents, 2007–2008. Journal of the American Medical Association,
Shields, K., et al. (2004). Barriers to physical activity: Qualitative data 303, 242−249.
on caregiver–daughter perceptions and practices. American Journal of Omar, M. A., Coleman, G., & Hoerr, S. (2001). Healthy eating for rural low-
Preventive Medicine, 27, 218−223. income toddlers: Caregivers' perceptions. Journal of Community Health
Hart, K. H., Herriot, A., Bishop, J. A., & Truby, H. (2003). Promoting Nursing, 18, 93−106.
healthy diet and exercise patterns amongst primary school children: A Paxton, A. E., Ammerman, A. S., Gizlice, Z., Johnston, L. F., & Keyserling,
qualitative investigation of parental perspectives. Journal of Human T. C. (2007). Validation of a very brief diet assessment tool designed to
Nutrition and Dietetics, 16, 89−96. guide counseling for chronic disease prevention. Abstract presented at:
Hart, L. G., Larson, E. H., & Lishner, D. M. (2005). Rural definitions for International Society for Behavior, Nutrition, and Physical Activity;
health policy and research. American Journal of Public Health, 95, June 2007; Oslo, Norway.
1149−1155. Perkins, M. B., Jensen, P. S., Jaccard, J., Gollwitzer, P., Oettingen, G.,
Henry, D. A. (2005). Nurse practitioner use of the transtheoretical model of Pappadopulos, E., et al. (2007). Applying theory-driven approaches to
change for breastfeeding facilitation. Clinical Excellence for Nurse understanding and modifying clinicians' behavior: What do we know?
Practitioners, 9, 192−194. Psychiatric Services, 58, 342−348.
Irwin, J. D., He, M., Bouck, L. M., Tucker, P., & Pollett, G. L. (2005). Perrin, E. M., Finkle, J. P., & Benjamin, J. T. (2007). Obesity prevention and
Preschoolers' physical activity behaviours: Parents' perspectives. the primary care pediatrician's office. Current Opinions in Pediatrics,
Canadian Journal of Public Health, 96, 299−303. 19, 354−361.
Jenkins, S., & Horner, S. D. (2005). Barriers that influence eating behaviors Powers, K.A., Rehrig, S.T., & Jones, D.B. (2007). Financial impact of
in adolescents. Journal of Pediatric Nursing, 20, 258−267. obesity and bariatric surgery. Medical Clinics of North America, 91,
Kang, H. T., Ju, Y. S., Park, K. H., Kwon, Y. J., Im, H. J., Paek, D. M., et al. 321–38, ix.
(2006). Study on the relationship between childhood obesity and Schwimmer, J. B., Burwinkle, T. M., & Varni, J. W. (2003). Health-related
various determinants, including socioeconomic factors, in an urban quality of life of severely obese children and adolescents. Journal of
area. Journal of Preventive Medicine and Public Health, 39, 371−378. the American Medical Association, 289, 1813−1819.
Kicklighter, J. R., Whitley, D. M., Kelley, S. J., Shipskie, S. M., Taube, J. Stanhope, M., & Lancaster, J. (1996). Community health nursing (4th ed.).
L., & Berry, R. C. (2007). Grandparents raising grandchildren: A St. Louis, MO: Mosby.
response to a nutrition and physical activity intervention. Journal of the Stettler, N., Signer, T. M., & Suter, P. M. (2004). Electronic games and
American Dietetic Association, 107, 1210−1213. environmental factors associated with childhood obesity in Switzerland.
Koplan, J. P., Liverman, C. T., Kraak, V. I., & Committee on Prevention of Obesity Research, 12, 896−903.
Obesity in Children and Youth (2005). Preventing childhood obesity: Stevens, A. B., Coon, D., Wisniewski, S., Vance, D., Arguelles, S., Belle,
Health in the balance: Executive summary. Journal of the American S., et al. (2004). Measurement of leisure time satisfaction in family
Dietetic Association, 105, 131−138. caregivers. Aging and Mental Health, 8, 450−459.
Lioret, S., Volatier, J. L., Lafay, L., Touvier, M., & Maire, B. (2009). Is food Story, M., Neumark-Sztainer, D., & French, S. (2002). Individual and
portion size a risk factor of childhood overweight? European Journal of environmental influences on adolescent eating behaviors. Journal of the
Clinical Nutrition, 63, 382−391. American Dietetic Association, 102, S40−S51.
Mausner, J. S., & Bahn, A. K. (1974). Epidemiology: An introductory text Sturm, R. (2005a). Childhood obesity—What we can learn from existing
(1st ed.). Philadelphia, PA: W.B. Saunders Company. data on societal trends, part 1. Preventing Chronic Disease, 2, A12.
McGarvey, E. L., Collie, K. R., Fraser, G., Shufflebarger, C., Lloyd, B., & Sturm, R. (2005b). Childhood obesity—What we can learn from exist-
Norman Oliver, M. (2006). Using focus group results to inform ing data on societal trends, part 2. Preventing Chronic Disease, 2, A20.
preschool childhood obesity prevention programming. Ethnicity and Taveras, E. M., Berkey, C. S., Rifas-Shiman, S. L., Ludwig, D. S., Rockett,
Health, 11, 265−285. H. R., Field, A. E., et al. (2005). Association of consumption of fried
National Cancer Institute, National Institutes of Health. (2005). Theory at a food away from home with body mass index and diet quality in older
glance: A guide for health promotion practice. Retrieved July 31, 2009, children and adolescents. Pediatrics, 116, e518−e524.
from: www.cancer.gov/theory. Weir, L. A., Etelson, D., & Brand, D. A. (2006). Parents' perceptions of
NC Prevention Partners. (2010). Starting the Conversation Series. Retrieved neighborhood safety and children's physical activity. Preventive
December 6, 2010, from: http://www.ncpreventionpartners.org/dnn/ Medicine, 43, 212−217.
services/resources/startingtheconversationseries/tabid/82/default.aspx. Wilfley, D. E., Tibbs, T. L., Van Buren, D. J., Reach, K. P., Walker, M. S.,
Nemet, D., Barkan, S., Epstein, Y., Friedland, O., Kowen, G., & Eliakim, A. & Epstein, L. H. (2007). Lifestyle interventions in the treatment of
(2005). Short- and long-term beneficial effects of a combined dietary– childhood overweight: A meta-analytic review of randomized controlled
behavioral–physical activity intervention for the treatment of childhood trials. Health Psychology, 26, 521−532.
obesity. Pediatrics, 115, e443−e449. Williams, J., Wake, M., Hesketh, K., Maher, E., & Waters, E. (2005).
Neumark-Sztainer, D., Story, M., Perry, C., & Casey, A. (1999). Factors Health-related quality of life of overweight and obese children. Journal
influencing food choices of adolescents: Findings from focus-group of the American Medical Association, 293, 70−76.
discussions with adolescents. Journal of the American Dietetic Wofford, L. G. (2008). Systematic review of childhood obesity prevention.
Association, 99, 929−937. Journal of Pediatric Nursing, 23, 5−19.