t:Perbhan R

Res

ic, Rter,

intervention is warranted. 2013 Elsevier Inc. All rights reserved.

Journal of Pediatric Nursing (2013) 28, 536547 Corresponding author: Sharon J. Tucker, PhD.DATA PUBLISHED IN 2010 suggest that 31.7% ofchildren 219 years old are overweight, defined as above the85th percentile for body mass index (BMI) with 16.9%above the 95th percentile (Ogden, Carroll, Curtin, Lamb, &Flegal, 2010). Overweight is associated with considerablehealth problems and is an important early risk factor for adultmorbidity and mortality. Diseases previously seen primarilyin adults such as diabetes, hypertension, heart disease,hyperlipidemia, reproductive problems, asthma and sleepdisorders are now being observed with overweight children(Lobstein, Baur, & Uauy, 2004). Pediatric primary careproviders, who usually have limited time, are often illequipped to approach these health concerns, and many times

do not have meaningful resource options for child and parentreferrals (Dorsey, Wells, Krumholz, & Concato, 2005; Talmi& Fazio, 2012). Alternative primary care models are thusneeded that leverage existing resources, bring appropriatetraining for providers, and offer children and parentsmeaningful interventions. The present study evaluated thefeasibility and outcomes of an innovative primary care modelfor addressing childhood obesity using a registered nurse(RN) to deliver health habits messages through motivationalinterviewing (MI) with 4-18-year old overweight childrenand their parents presenting for a scheduled well-child visit.

Background

Factors contributing to overweight and obesity are(controls 1.9, intervention 4.6). Intervention effects were found for self-reported daily fruit/vegetableconsumption, physical activity, and screen time. Satisfaction was high. Further study of the PC nursing0hChildhood overweight;Primary care;Nurse led intervention;Pediatric obesity;Motivational interviewingNursing Research, Mayo Clinic, Rochester, MNePediatrics Research and Informatics, Mayo Clinic, Rochester, MN

Key words:The purpose of this study was to test the Let's Go 5-2-1-0 program delivered through motivationalinterviewing by nurses with 4-18-year-old overweight children and parents in primary care (PC). Aquasi-experimental design allocated 60 control families to standard clinical care (SCC) and 70 familiesto SCC plus the 5-2-1-0 intervention. Drop-out rates were 9 and 35% at 6 months and 25 and 41% at12 months, respectively for control and intervention participants. BMI percentile trended (p = .057)toward decline (M change = 3.0 versus 1.5) for intervention children at 6 months (n = 52 control, 44intervention), and nonsignificantly (p = 0.14) for both groups (43 control, 40 intervention) at 12 monthsReducing Pediatric OverweighMotivational Interviewing inSharon J. Tucker PhD, RNa,, Karen L. YttTammy L. Schmit RN, BSNb, Debra I. MucChristine M. Lohse MSc, Christine M. AustiaDepartment of Nursing Services and Patient Care, NursingUniversity of Iowa Hospitals & Clinics, Iowa City, IAbCommunity Pediatrics and Adolescent Medicine, Mayo ClincBiomedical Statistics and Informatics, Mayo Clinic, RochesdE-mail address: sharon-tucker@uiowa.edu.

882-5963/$ see front matter 2013 Elsevier Inc. All rights reserved.ttp://dx.doi.org/10.1016/j.pedn.2013.02.031rimary Careerg MDb, Lisa M. Lenoch BSN, RNb,RN, CNPb, Judith A. Wooten RNb,N, ADd, Kari J. Mongeon Wahlen RN, MSe

earch and Evidence-Based Practice,

ochester, MNMNNurse-Deliveredcomplex and often oversimplified (McAllister et al., 2009).

Nonetheless, there are identified behavioral components Another group launched a larger multi-site approach to

537Reducing Pediatric Overweightassociated with obesity that providers can target such asadequate physical activity, limiting screen time, eating fruitsand vegetables, and limiting soft drink consumption.Moreover, recommendations from the American Academyof Pediatrics (AAP) for health providers include discussinghealth habits, not body habitus, as part of efforts to controlobesity (Krebs et al., 2003, reaffirmed 2006). The researchon such approaches or treatment and prevention approachesbased in primary care are limited (Kamath et al., 2008;McGovern et al., 2008; Waters, Swinburn, Seidell, & Uauy,2010; Whitlock, O'Connor, Williams, Beil, & Lutz, 2010).

One approach that appears promising as a primary carestrategy for helping children and parents with adoptingpositive health habits is motivational interviewing (Ritchie,Crawford, Hoelscher, & Sothern, 2006). Motivationalinterviewing (MI) has its roots in addictions therapies andwas initially developed for problem drinking by WilliamMiller (Miller, 1983). Drawing from the transtheoreticalframework and stages of change theory, it is a client-centered, directive method for enhancing intrinsic motivationto change select behavior by exploring and resolvingambivalence (Miller & Rollnick, 2002).

A systematic review of the effects of MI for pediatricobesity was conducted by Resnicow, Davis, and Rollnick(2006). They concluded that MI might be feasible withchildren and adolescents, yet the data were insufficient todetermine efficacy as a behavioral strategy for prevention ortreatment of pediatric obesity. Suarez and Mullins (2008)found similar conclusions in a systematic review of outcomestudies of health behavior change interventions using MIwith pediatric populations. They remarked that MI aimed ataffecting a variety of clinical health conditions appearsfeasible as a stand-alone or adjunctive treatment, howevertoo few quality studies had yet been conducted.

Schwartz et al. (2007) examined whether pediatricians anddietitians would implement an office-based obesity preventionprogram using motivational interviewing as the primaryintervention for children at the 8595th BMI percentile orhad one parent with a BMI of 30 or greater. Using anonrandomized design, 15 pediatricians from 15 differentpediatric practices were recruited to participate in the study andassigned to one of three groups with five clinics per group:control group (standard care), minimal intervention (onepediatrician visit only), or intensive intervention (one pediatri-cian and one dietician visit). Five dieticians from theirrespective practices participated in the intensive interventiongroup only. Children and their parents were enrolled at each siteinto their respective groups: control group (n = 21), minimalintervention (n = 30), and intensive intervention (n = 30).Nonsignificant decreases in BMI at 6-month follow-up werefound for all three groups with the lowest change in the controlgroup and the greatest change in the intensive dose group.Parent satisfaction was high. They concluded that MI lookspromising for primary care settings but needs further studygiven the small sample and method and dosage limitations.examine if providers would adopt MI into their practices(Polacsek et al., 2009). These researchers and cliniciansincorporated MI as a strategy for delivering the Let's Go 5-2-1-0 Program, formerly known as the Keep Me Healthy 5-2-1-0 Program. This program was developed as part of thePromoting Healthy Habits for Life Maine Youth OverweightCollaborative (MYOC) located in Maine and includes fourprimary health habit targets 5-2-1-0 (Figure 1) that helpchildren and parents focus on a behavior change they areready to make.

A framework and a toolkit were developed to assistproviders in helping patients adopt healthy behaviors. The 5-2-1-0 program was initially evaluated through an experi-mental field trial in 12 urban and rural pediatric practices inMaine. Findings indicated improvements in provider prac-tices and suggested promise for the 5-2-1-0 as anintervention for addressing overweight risk among childrenand youth in pediatric primary care (Polacsek et al., 2009).The Centers for Disease Control and Prevention (CDC)highlights the 5-2-1-0 program as one effective strategy foraddressing overweight in young patients (CDC, 2011).

Published child outcomes from the Let's Go 5-2-1-0 program are limited but growing. The widespread adoptionefforts in the Portland, MA area have been evaluated throughparent telephone screens, school surveys, and chart reviewsin health clinics. A 2011 report (Let's Go, 2011) indicatesthat between 2007 and 2011, the number of parents able tocorrectly identify all four 5-2-1-0 recommendations in-creased by 34%, child's adherence to the 5 (five dailyservings of fruits and vegetables) increased by 63%, andchild's adherence to 2 (2 hours or less of daily recreationalscreen time) increased by 18%. A review of patient charts forheight and weight showed the prevalence of overweight andobesity among children decreased non-significantly from33% in 2006 to 31% in 2009, and the proportion of 35 yearold females who were overweight and obese declinedsignificantly from 31% in 2006 to 25% in 2009. Similarfindings were reported by Tucker et al. (2011) who studiedthe Let's Go approach with fourth to fifth graders and foundsignificant effects on objectively measured child physicalactivity levels and self-reported health habits (5-2-1-0 targets)following a 4-month intervention delivered by nursingstudents at the school setting.

The duration (dosage) of the program needed for positiveoutcomes is not yet known and a gap remains in this area formany behavioral health interventions (Daniels, Jacobson,McCrindle, Eckel, & Sanner, 2009). However, others havereported that it appears a 46 month time for behavioralinterventions is needed for positive outcomes (Cowburn,Hillsdon, & Hankey, 1997; Oude Luttikhuis et al., 2009).

This paper reports findings of a study that evaluated thefeasibility and BMI and healthy habits outcomes of a primarycare intervention that leveraged the RN role to deliver the 5-2-1-0 health habits messages through MI over 6 months with418 year old overweight (BMI = 8595th percentile)

Children and their parents were recruited from a

etab

scho

ever

ilk i

Let's

538 S.J. Tucker et al.Midwestern pediatric community practice in an academicmedical center. To be eligible for the study, children had tobe 418 year olds, at the 8595th BMI percentile, andpresenting for a well-child visit. Children with significantco-morbidities and those presenting for acute care concernswere excluded to reduce confounding factors and to avoidchildren and their parents. Study aims were to (a) comparechange in BMI percentile and self-reported health habits ofchildren and parents between those who received standardcare (SC) and those who received SC plus the MIintervention, and (b) describe parent, child and providersatisfaction with the MI intervention.

Method

Design

This quasi-experimental study evaluated feasibility andlogistics of implementing an RN delivered MI intervention inwell-child care, and estimated preliminary effects of theintervention. The study involved two phases. Phase 1included enrollment of control participants, development ofa standardized protocol for the brief MI intervention, andtraining and practice for the RNs who delivered the MIintervention in phase 2. Phase 2 involved enrollment ofintervention participants and delivery of the 5-2-1-0 inter-vention. Parents and children in phase 1 received SC, whileparents and children in phase 2 received SC plus a structuredMI intervention integrating 5-2-1-0 delivered by an RN.Random assignment was waived to avoid contamination ofgroups once the intervention was introduced.

Setting and Sample

5 = Eat at least five servings of fruits and veg

2 = Limit TV or computer use (not related to

1 = Get one hour or more of physical activity

0 = Drink less sugar. Try water and low fat m

Figure 1 5-2-1-0interference with treating acute problems. Additionalexclusion criteria included: (1) taking any type of birthcontrol including Depo Provera injections, oral steroids,seizure medications, antidepressants, or medications thataffect appetite; (2) pregnancy at time of enrollment orduring the study; (3) a diagnosis of any mental healthproblems (Autism, ADHD, MR etc.); (4) following anyspecial diets; and (5) non-English speaking or significantlanguage barriers.Power EstimateTo evaluate preliminary effects, a power analysis was

conducted using BMI percentile as the primary endpoint toestimate sample size. Estimates in published studies havereported large standard deviations for BMI making thesample sizes needed for detecting effects very large. Thus, anarrower group of participants, those whose BMI fellbetween the 85th and 95th percentile, was chosen to limitthe overall standard deviation. A sample size of 51participants in each group provided 80% power to detect amean difference between groups of 2.5 percentage pointsand a standard deviation of 5.0 percentage points based on aone-tailed test with alpha set at 0.05. To account forexpected attrition, 60 control and 70 intervention partici-pants were enrolled.

Variables and Measures

BMI, BMI percentile and healthy habits were the primaryvariables of interest in terms of preliminary interventioneffects. BMI percentile was determined using CDC growthcharts that plot BMI along age and gender specificdevelopmental percentiles. In this study, BMI percentilewas auto-calculated within the electronic medical record.

Healthy habits were measured by the age specific HealthyHabits Survey, developed by the Maine Youth OverweightCollaborative (Maine Center for Public Health, 2012;Polacsek et al., 2009). This survey, designed specificallyfor the 5-2-1-0 Program, is a 10-item measure of healthhabits related to nutrition, screen time, physical activity, andfamily eating patterns. Six items ask respondents to ratefrequency to which they consume or engage in the item, twoitems are yes/no responses, and two include multiple choiceresponse options. There are two versions of this survey thatare the same content but worded for who is completing thesurvey, based on the age and development of the child. TheAges 29 version is worded for parents to complete for the

les a day.

ol) to two hours or less a day.

y day.

nstead of soda and drinks with lots of sugar.

Go key messages.child and the Ages 1018 version is worded for self-completion by the child. These versions were usedaccordingly in this study, with parents completing if childrenwere under 10, and children completing alone or with parentsif over age 9. There is no composite score for this measure;rather each item is evaluated separately. No psychometricdata are available for this measure.

Satisfaction with the approach was measured usingresearcher developed evaluation surveys designed to be

completed by intervention children/parents and providers. was used as the foundation for the protocol, with specific

539Reducing Pediatric OverweightChildren and parents were asked to complete the survey atthe 6-month visit when the intervention ended, to rate theoverall program and the different strategies for engagingthem. The 12-month follow-up survey was also administeredto participants, to gather more comprehensive feedbackabout overall satisfaction with the program and perceivedbenefits over time. The 6-month provider survey askedproviders to rate their overall opinion of the program, theenrollment process, and the impact of the program on patientflow. As these tools were developed for this study, nopsychometric data were available.

Standard Clinical Care

Both groups were offered SC during their well-child visit.SC included providing a print out and review of BMI andBMI percentile. This outlines where the child standscompared to other children of the same gender, age, heightand weight. Standard hand-outs related to healthy weight andmanagement were provided by the clinical staff, either RN orprovider, and referrals were made as needed for children inthe overweight and obese BMI categories. Referrals primarilywere for dietary or endocrinology evaluations as part of SC.

Motivational Interviewing Intervention

The intervention was modeled after the Let's Go 5-2-1-0 Program. This program was developed around the fourspecific key messages (Figure 1) that are delivered usingmotivational interviewing (MI) techniques. As part of MIprinciples, readiness to change is viewed not as a patient trait,but a fluctuating product of interpersonal interaction. Thetherapeutic relationship is viewed as a partnership wherebythe interventionist does not prescribe specific methods ortechniques and patients are responsible for their progress.Interventionist strategies include the following:

Expressing acceptance and affirmation; Eliciting and selectively reinforcing the client's own selfmotivational statements of problem recognition, concern,desire and intention to change, and ability to change;

Monitoring the client's degree of readiness to change; Ensuring that resistance is not generated by jumpingahead of the client; and

Affirming the client's freedomof choice and self-direction.

Training began by having the three interventionists attend a3-day MI workshop offered offsite by an MI expert. On-sitetraining was then provided to all study team members by thesame MI expert, who was a PhD prepared child psychologistwith substantial training and experience along with certifica-tion as a MI trainer. The onsite training including didacticcontent, practice scenarios, and guidance in development of thebrief MI intervention protocol. The Let's Go 5-2-1-0 Programmaterials and parent handouts. For example, the HealthyLifestyle Goal SettingWorksheet asked participants to select afocus among the 5-2-1-0 target areas, and the Readiness Rulerasked participants to rate on a 110 scale their readiness tomake changes according to their selected focus and goalcreated for the focus. Standardized program handouts forparents and children included strategies to meet goals for theselected target area. Cue cards were developed for eachintervention session to guide the interventionist to be consistentin delivery with all families. Three RN interventionists, allclinical staff, were trained to offer the intervention. However,due to other clinical demands, only one RN providedN 90% ofthe intervention. The other two were pulled away from theproject after the initial 12 months of enrolling given otherclinical demands and need to maximize existing resources.

Study Procedures

Human Subjects ProtectionReview and approval of human subjects research were

received from the Institutional Review Board (IRB) prior tolaunching any study aspects. Based on the agency IRBpolicies, children under 8 years old were not required toprovide written assent; children 812 years of age providedassent on a signed assent form alongwith signed consent fromparents; and children over 12 provided written assent by co-signing the consent form with their parents. Assent involvedexplaining to the children in very simple terms what the studyincluded and what was expected of them. The choice to notparticipate without anyone being mad was made clear, withthe study member signing the assent form indicating the childhad given their assent to participate in the study.

Phase 1 ProceduresPhase 1 was launched and eligible control participants

were enrolled in the study. Recruitment of participantsoccurred through involvement of the clinical staff. When achild and parent presented for a well-child visit, they wereassessed for meeting study eligibility criteria by clinical staff.If eligible, the parent was asked if they would be interested inlearning about the study and if so, a flag was noted on theroom door and a study nurse or assigned study coordinatorwas paged. This study team member then approached thechild and parent to discuss the study and enroll the family asappropriate. Parents were consented and children assented,and data were collected by the study member on child BMI,demographics, and health habits. As part of SC at thispediatric primary care practice, control participants wereoffered a review of their BMI and BMI percentile, standardhand-outs related to healthy weight management, andreferrals as needed. Data collected at baseline were collectedagain 6 and 12 months later at the clinic by the studycoordinator assigned to the study. Participants were offered apedometer, water bottle, timer, or jump rope along withparking passes as incentives for returning for the follow up

visits. During this phase the intervention was also established

juice, fruit or sports drinks, or non-diet soda or punchper day; and

540 S.J. Tucker et al.into a study protocol and the interventionists were trained inthe protocol.

Phase 2 ProceduresPhase 2 was launched after control participant accrual was

met and the intervention protocol was established and pilotedwith a non-study patient including the follow-up telephonesessions. Intervention parents and children were enrolledusing the procedures established for control participants withthe study coordinator or study nurse collecting data and thestudy nurse implementing the MI intervention. After datawere collected, the first MI intervention session wasdelivered the same day or within the next week by the RNinterventionist using the MI approach, techniques and skillsdescribed above. During the first MI session (~30 minutes)participants completed the Readiness Ruler and HealthyLifestyle Goal Setting Worksheet described earlier. Theythen discussed strategies for working on the goal selected,reviewed the weekly phone follow-up regimen for the next4 weeks, and scheduled the next onsite visit that occurredabout 1 month later. After the second onsite MI session,periodic phone sessions were scheduled with the frequencyagreed upon by the parent, child and interventionist.

The telephone sessions were structured to review goals,problem solve barriers, and discuss strategies for meetingestablished goals. The MI skills used in the onsite sessionswere used in the telephone sessions. Other tools to identifyindividualized strategies included handouts created by the 5-2-1-0 Program. The interventionist consulted with a clinicaldietician when dietary/nutrition issues extended beyond thescope of the RN practice. Other Web sites were used asappropriate to provide additional ideas (e.g., the CDC Website) for parents and children.

Six and 12 months after enrollment, BMI and healthhabits data were again collected by the study staff.Additionally, an evaluation survey was completed byintervention participants (parents and child if appropriate)at both follow-up visits. At the 6-month visit, the data werefirst collected and a booster MI session was then provided bythe interventionist; and referrals for further services weremade as indicated. As with control participants, interventionparticipants were offered pedometers, water bottles, jumpropes, and parking passes as incentives for returning for thefollow up appointments and assessments. Providers werealso asked to evaluate their satisfaction with the program6 months after the intervention participants were enrolled.

Weekly study team meetings were held to evaluateprogress and trouble shoot study issues. To promote andevaluate intervention fidelity, MI sessions were recorded(with parent consent) and select sessions were reviewed bythe study team members and the MI trainer with feedbackprovided to the interventionist on how to strengthen theintervention and MI strategies and skills. The first twosessions were purposefully selected and then two additionalsessions were picked by the interventionist. A decrease in the number of servings of whole milk perday, and an increase in the number of servings of nonfator reduced fat milk per day.

The numbers of participants who maintained or achievedfive servings of fruits or vegetables per day and removed aTV or PC from their room or did not have a TV or PC in theirroom to begin with were also evaluated.

Differences between participants in the control andintervention groups were evaluated using Wilcoxon ranksum for ordinal or interval data and chi-square and Fisher'sexact tests for categorical data. Statistical analyses wereperformed using the SAS software package (SAS Institute,Cary, NC). All tests were two-sided and p-values b .05 wereconsidered statistically significant. Participant (parent, childand provider) evaluation feedback was summarized aspercentages or means for items that asked for a rating.Themes were drawn from the narrative responses to theopen-ended questions on the evaluation surveys.

Results

Demographic Characteristics

Figure 2 outlines the participant enrollment and studycompletion numbers. Ninety-six participants (74%) whoenrolled in the study completed the 6-month assessment, 52in the control group and 44 in the intervention group.Demographic characteristics (Table 1) for these 96 partici-pants did not differ significantly between groups except forprimary person responsible for parenting (p = .021), whereboth parents was reported less frequent and mothersmore frequent in the control group. Eighty-three participantscompleted the 12-month assessment: 43 in the control groupand 40 in the intervention group.Data Analysis

Demographic data were analyzed using descriptivestatistics. Changes in BMI, BMI percentile, and the itemson the Healthy Habits Survey from the baseline to either the6 or 12 month visit were evaluated both as continuousvariables and using indicator variables (yes/no) for positivechanges. Positive changes were defined as follows:

A decrease in BMI percentile; An increase in the number of servings of fruits orvegetables per day;

A decrease in hours of TV, movies, video, or computergames per day;

An increase in hours of active play per day; A decrease in the number of servings of 100% fruit

Ased y (N

dom130)

541Reducing Pediatric OverweightAssesEligibilit

Allocation to Control Group Became

Non Ran(N = Study Aim 1: Compare Change in BMI Percentile andSelf-Reported Health Habits of Children and ParentsBetween Those Who Received SC and Those WhoReceived SC Plus the MI Intervention

Body Mass IndexBMI and responses to the Healthy Habits Survey at the

baseline, 6 month, and 12-month visit are summarized inTables 2 and 3, for control and intervention groupsrespectively. Baseline dependent variables did not differbetween groups except for self-reported hours of active play

ANALYZED (n = 52) 6-Month BMI & Health Habits (n = 52)

12-Month BMI & Health Habits (n = 43)

A

Analysis

Standard Care (n = 57) All received standard care

StanInte

LOST TO FOLLOW-UP 6-month follow-up (n = 5)

Lost/Dropped

12-month follow-up (n = 9) Lost/Dropped, Withdrew

L

Follow-Up

n = 60 Ineligible (n = 3)

Figure 2 Enrollment and pllocation to Intervention Group

for= 721)

Became

ized

Excluded (n = 591) BMI < 85% (n = 328)BMI > 95% (n = 97) Medication exclusion (n = 18)Declined (n = 94) Not retained (n = 27) Health exclusion-psych, ADHD,diabetes - the rest were blank (n = 7) 10 non-English (n = 10)10 - not indicated (n = 10)(p = .002), control participants reported M = 2.0 (Mdn = 2;range = 0.56) and intervention participants reported M =1.3 (Mdn = 1; range = 0.254.5).

From baseline to the 6-month visit, 52 control and 44intervention families provided full datasets. Among theseparticipants, the BMI percentile decreased, although thegroup differences only trended towards statistical signifi-cance (mean decreases of 1.5 and 3.0, control andintervention groups respectively; p = .057). Twenty-three(44%) of control children decreased their BMI percentilecompared with 27 (61%) intervention children (p = .094).

NALYZED (n = 44) 6-Month BMI & Health Habits (n = 44) 6-Month Participant Evaluation (n = 35)

12-Month BMI & Health Habits (n = 40) 12-Month Participant Evaluation (n = 21)

dard Care Plus Motivational rviewing Intervention with 5-2-1-0 (n = 68)

Received full allocated intervention (n = 40) Received partial allocated intervention (n = 24) Did not receive any allocated intervention (n = 4)

Time constraints

n = 70

OST TO FOLLOW-UP 6-month follow-up (n = 24)

Dropped, death in family, relocation

12-month follow-up (n = 4) Dropped, death in family, divorce, relocation

Ineligible (n = 2)

articipation flowchart.

Table 1 Family Demographics for 96 Participants.

Control (n = 52) Intervention (n = 44)

p-ValueFeature Mean Median Range Mean Median Range

Age of participant 9.5 9.5 415 9.9 9 418 .82Age of parent (n = 94) 38.2 38 2054 40.1 39 2752 .24Missing data (n = 1 control, 1 intervention)

n % n % p-Value

Parent completing form .38Mother 45 (87) 36 (82)Father 7 (13) 6 (14)Other 0 2 (5)Parental marital status (n = 94) .60Married 36 (72) 37 (84)Single 5 (10) 3 (7)Living with a partner 4 (8) 2 (5)Divorced 5 (10) 2 (5)Missing data (n = 2 controls) 1 1Primary person responsible for parenting .021Both mother and father 30 (58) 34 (77)Mother 19 (37) 8 (18)Father 3 (6) 0Grandparent 0 1 (2)Other 0 1 (2)Parental ethnicity .73Caucasian 42 (81) 38 (86)Asian 2 (4) 3 (7)Black/African American 3 (6) 2 (5)Hispanic/Latino 1 (2) 1 (3)Multi-ethnicity 2 (4) 0Other 2 (4) 0Parental education level .60Less than high school 3 (6) 2 (5)High school or GED 4 (8) 0Some college or vocational training 11 (21) 9 (20)2-Year college degree 5 (10) 5 (11)4-Year college degree 16 (31) 14 (32)Post-baccalaureate degree 13 (25) 14 (32)Parental employment (n = 93) .32Full-time 28 (54) 25 (61)Part-time 11 (21) 11 (27)Not employed outside of home 13 (25) 5 (12)Missing data (n = 3 intervention)Household annual gross income ($, n = 95) .48020,000 5 (10) 3 (7)20,00130,000 5 (10) 1 (2)30,00140,000 2 (4) 1 (2)40,00150,000 4 (8) 3 (7)50,00160,000 6 (12) 7 (16)60,00170,000 4 (8) 1 (2)70,00180,000 6 (12) 3 (7)80,001+ 19 (37) 25 (57)Missing data (n = 1 control) 1 0Gender of child .31Female 23 (44) 24 (55)Male 29 (56) 20 (45)

542 S.J. Tucker et al.

Thirty (58%) children in the control group either maintainedor decreased their BMI percentile compared with 32 (73%)children in the intervention group (p = .12). A largerproportion of participants in the intervention groupdropped/withdrew from the study at 6 months (24 out of68; 35%) compared with only 5 (9%) participants in thecontrol group (p b .001).

From baseline to the 12-month visit, 43 control and 40intervention families provided full datasets. Among thesefamilies, BMI percentile decreased for both the control andintervention groups with a larger decline in the interventionparticipants, although not significantly different (meandecreases of 1.9 and 4.6, respectively; p = .14). A

68; 41%) compared with 14 (25%) of the 57 participants inthe control group (p = .050).

Healthy HabitsHealthy habits survey data are presented over time for

each group in Tables 2 and 3. For the families whoprovided full datasets, we observed a significant increasein the number of servings of fruits and vegetables perday from baseline to 6-month follow-up among theintervention families compared with the control families(mean increases of 1.2 and 0.1, respectively; p b .001). Agreater proportion of participants in the interventiongroup either maintained or achieved five servings of

eline

ge

28.25

Table 1 (continued)

Control (n = 52) Intervention (n = 44)

p-ValueFeature Mean Median Range Mean Median Range

n % n %

Ethnicity of child (n = 93) .94Caucasian 40 (80) 36 (84)Asian 2 (4) 2 (5)Black/African American 3 (6) 2 (5)Hispanic/Latino 1 (2) 1 (2)Multi-ethnic 2 (4) 2 (5)Missing data (n = 2 controls, 1 intervention) 3 2Other 2 (4) 0

543Reducing Pediatric Overweightgreater proportion of participants in the intervention groupdropped/withdrew from the study at 12 months (28 out of

Table 2 Control Group BMI and Healthy Habits Survey at Bas

Feature

Baseline (n = 52)

Mean(SD) Median Ran

BMI 21.0 (2.7) 21.0 17.2BMI percentile 90.7 (2.8) 91.0 859Servings of fruits or vegetables perday (n = 94)

2.6 (1.4) 2.5 06

Dinners with family per week 5.4 (1.8) 6.0 07Breakfasts per week 6.2 (2.0) 7.0 07Fast food per week 1.5 (1.3) 1.0 07

Hours of TV per day (n = 95) 2.4 (1.5) 2.0 08Hours of active play per day (n = 90) 2.0 (1.2) 2.0 0.56Servings of 100% fruit juice per day 0.7 (1.2) 0.0 08Servings of fruit/sports drinks per day 0.5 (0.7) 0.0 03Servings of soda/punch per day 0.7 (1.3) 0.0 08Servings of water per day 2.8 (1.8) 3.0 08Servings of whole milk per day 0.6 (1.8) 0.0 08Servings of nonfat or reduced fatmilk per day

2.1 (1.3) 2.0 04

n %

TV in room 14 27PC in room (n= 95) 3 6fruit or vegetables per day compared with the controlgroup (30 versus 10%; p = .021).

, 6 Month, and 12 Month Visit.

6 Months (n = 52) 12 Months (n = 43)

Mean(SD) Median Range

Mean(SD) Median Range

21.3 (2.9) 21.8 16.928.0 21.5 (3.1) 21.2 16.628.889.2 (6.9) 91.0 5598 88.7 (6.1) 91.0 68972.9 (1.6) 3.0 07 3.0 (1.4) 3.0 08

5.3 (1.9) 5.75 07 5.5 (1.9) 6.0 075.9 (1.8) 7.0 27 6.3 (1.5) 7.0 171.3 (1.0) 1.0 05 1.2 (1.0) 1.0 03

2.2 (2.0) 2.0 0.55 2.0 (1.1) 1.5 0.551.8 (1.2) 1.5 0.164.5 1.6 (1.2) 1.5 0.350.7 (1.2) 0.5 08 0.6 (0.7) 0.0 030.4 (0.6) 0.0 02 0.3 (0.7) 0.0 030.5 (1.0) 0.0 05 0.3 (0.5) 0.0 023.2 (1.7) 3.0 08 3.0 (1.8) 3.0 0100.5 (1.2) 0.0 05 0.3 (0.9) 0.0 042.0 (1.3) 2.0 06 2.1 (1.0) 2.0 03

n % n %

20 39 14 348 16 6 15

Table 3 Intervention BMI and Healthy Habits Survey at Baseline, 6 Month, and 12 Month Visit.

Feature

Baseline (n = 44) 6 Months (n = 44) 12 Months (n = 40)

Mean(SD) Median Range

Mean(SD) Median Range

Mean(SD) Median Range

BMI 21.6 (3.0) 21.0 17.028.2 21.5 (3.2) 21.0 15.928.4 21.6 (3.3) 21.4 16.129.3BMI percentile 90.9 (3.0) 91.0 8595 87.9 (6.7) 89.5 6696 86.2 (8.9) 89.0 5797Servings of fruits or vegetables perday (n = 94)

2.8 (1.3) 3.0 16 4.0 (1.4) 4.0 1.510 3.8 (1.1) 4.0 1.57

Dinners with family per week 5.7 (1.9) 7.0 07 5.8 (1.8) 6.75 07 6.1 (1.6) 7.0 07Breakfasts per week 6.4 (1.5) 7.0 27 6.5 (1.2) 7.0 27 6.5 (1.1) 7.0 1.57Fast food per week 1.3 (0.8) 1.0 04 1.1 (0.7) 1.0 03 1.0 (1.1) 1.0 05Hours of TV per day (n = 95) 2.1 (1.4) 2.0 0.59 1.7 (1.1) 1.5 0.256 1.5 (1.0) 1.5 0.254

4.5

.5

p s, p ts, p

544 S.J. Tucker et al.Looking at healthy habits data at baseline to 12-month

Hours of active play per day (n = 90) 1.3 (0.9) 1.0 0.25Servings of 100% fruit juice per day 0.6 (0.8) 0.0 04Servings of fruit/sports drinks per day 0.4 (0.7) 0.0 03Servings of soda/punch per day 0.4 (0.8) 0.0 03Servings of water per day 3.4 (2.5) 3.0 012Servings of whole milk per day 0.1 (0.5) 0.0 03Servings of nonfat or reduced fatmilk per day

2.3 (1.4) 2.0 06

n (%)

TV in room 13 30PC in room (n = 95) 3 7

Significant difference between control and intervention participants, Significant difference between control and intervention participant Significant difference between control and intervention participanfollow-up for the families with full datasets, we observed anumber of differences between groups. First, we observeda significant increase in the number of servings of fruitsand vegetables per day in the intervention group comparedwith the control group (mean increases of 1.0 and 0.1,respectively; p b .001). A greater proportion of participantsin the intervention group increased the number of servings offruits and vegetables per day compared with the controlgroup (75 versus 33%; p b .001). A greater proportion ofparticipants in the intervention group decreased the hours ofTV watched per day compared with the control group (63versus 39%; p = .035). In addition, a greater proportion ofparticipants in the intervention group increased the hours ofactive play per day compared with the control group (61versus 27%; p = .004).

Table 4 Changes in BMI and Healthy Habits Survey From Baseline

Feature

Control (n = 57)

Mean Median Rang

Change in BMI 0.3 0.3 3.1 toChange in BMI percentile 1.4 0.0 32 to 6

n %

Decrease in BMI 19 33Decrease in BMI percentile 23 40Intent-to-Treat Analysis

1.3 (0.7) 1.0 0.253.5 1.6 (1.3) 1.5 0.2570.5 (0.6) 0.0 02 0.4 (0.5) 0.0 01.50.2 (0.4) 0.0 02 0.1 (0.3) 0.0 010.2 (0.4) 0.0 02 0.3 (0.6) 0.) 02.53.7 (1.7) 3.5 1.58 3.7 (1.6) 3.75 18

0.0 (0.2) 0.0 01 0.1 (0.3) 0.0 022.5 (1.3) 2.5 06 2.4 (1.2) 2.5 05

n (%) n (%)

12 27 8 204 9 2 5

.05..001. .01.Given the concerns ofmissing data and drop-outs differencesbetween control and intervention families, for the primaryendpoint of BMI percentile we performed an intent-to-treatanalysis, which compares groups based on initial allocationrather than what was actually administered. Assumingparticipants who withdrew from the study did not make anychange in BMI or BMI percentile, the resulting changes fromthe baseline to the 6 month visit are summarized in Table 4.Mean change in BMI percentile was 1.4 for control childrenand 2.0 for intervention children. Similar findings wereobserved for the baseline to 12-month outcomes, againassuming that participants who withdrew from the study hadno change in BMI or BMI percentile. Findings (summarized inTable 5) indicate that mean change for BMI percentile was1.4for control children and2.7 for intervention children (p = .46).

to 6 Month Visit for 125 Participants in Intent-to-Treat Analysis.

Intervention (n = 68)

p-Valuee Mean Median Range

2.4 0.0 (0.0) 2.2 to 2.5 .0502.0 (0.0) 21 to 5 .15

n % p-Value

23 34 .9527 40 .94

and a trend for declining BMIs for both groups over time.

Table 5 Changes in BMI and Healthy Habits Survey From Baseline to 12 Month Visit for 125 Participants in Intent-to-Treat Analysis.

ange

.8 to9 to 6

545Reducing Pediatric Overweight2. Study Aim 2: Describe Parent, Child and ProviderSatisfaction With the MI Intervention

ParentChild 6-Month EvaluationAt the 6-month follow-up, 35 participants returned the

completed evaluation. Parent/child ratings (n = 35) of theoverall program ranged from 6 to 10 (10 being the best) withan average of 8.8 (SD = 1.11). Satisfaction ratings of thetypes of participation strategies ranged from 6 to 10 with anaverage of 8.9 (SD = 1.16). In terms of satisfaction withfrequency of the phone calls using a 13 rating (1 = toomany, 2 = just right, 3 = too few), 33 (94%) participantsindicated that the frequency was just right, one indicated toomany, and another person commented there might be toomany initially. Forty-five percent of parents identifiedmaking the most change related to five fruits and vegetablesper day (n = 16), followed equally at 20% each by 2 hoursor less of screen time (n = 7) and 1 hour or more physicalactivity per day (n = 7). Five (14%) parents identifiedmaking change for 0 drink less sugar.

Children also reported where they made the most change,with more than one target area chosen frequently. Fifty-fourpercent identified they made change for five fruits andvegetables per day (n = 19), 37% identified they madechange for 1 hour or more physical activity per day (n =13), 31% identified change for 2 hours or less of screentime (n = 11), and 23% indicated change for 0 drink lesssugar (n = 8). Parent/child dislikes of the program did notcluster around any particular theme except the commute to theclinic and overall time commitment to the program aspects.Eighty percent of participants recommended the programcontinue and only one person indicated they would not.

Feature

Control (n = 57)

Mean Median R

Change in BMI 0.5 0.0 1Change in BMI percentile 1.4 0.0 1

n %

Decrease in BMI 15 26Decrease in BMI percentile 24 42ParentChild 12-Month EvaluationTwenty-one (48%) of the 44 intervention participants who

completed the 6-month intervention returned the 12-monthevaluation survey. All 21 indicated they would participateagain in the program, and 90% would recommend theprogram (2 did not answer the question). Participantsidentified a number of challenges between the 6- and 12-month visits where they no longer received regular phonecalls and struggled with slipping to previous patterns. Othersidentified they were keeping up with the changes and felt theearlier frequent meetings helped reinforce learning points.Group differences were not significant for BMI or BMIpercentile; however reductions were larger for interventionparticipants. The reported health habits of children weresignificantly improved at the 12-month follow-up, ascompared to the control group. More intervention childrenand/or parents reported increases in servings of fruits andvegetables and hours of active play per day, along withsignificant decreases in hours of television watched per day.These findings are consistent with program effects reportedin the Let's Go 2011 report (Let's Go, 2011) with regard tochanges in servings of fruits and vegetables and hours ofscreen time, as well as trends for declining BMI percentile.They are also consistent with the outcomes reported byProvider 6-Month EvaluationTen providers and two nursing staff returned evaluation

surveys. Provider satisfaction ratings of the overall programand of the enrollment process averaged 8.6 with a range of 710 (scale of 110with 10 the best). Nurses' ratingswere 10 forboth elements. Regarding impact on patient flow of activity(scale of 110 with 10 being the best), average provider ratingwas a 9 and average nurse rating a 5. All 12 ratersrecommended continuing the program. Providers noted theyliked the care access for families, the concrete ideas forchange, involvement of children, patient/family centeredconcepts, the augmenting of provider care, a structured weightcontrol program for motivated patients, and a prevention tool.

Discussion

Findings from the study include several significantchanges in health habits among the intervention participants

Intervention (n = 68)

p-ValueMean Median Range

3.5 0.1 0.0 3.6 to 3.1 .082.7 0.0 33 to 4 .46

n % p-Value

19 28 .8425 37 .54Tucker et al. (2011) regarding a school-based approach thatintegrated the Let's Go 5-2-1-0 approach.

The study findings suggest that behavior change takes timeand short-term approaches are not as likely to be sustainable(Passehl et al., 2004; Schwartz, 2010). Intervention participantsreportedmore positive changes in health habits at the 12-monthassessment, and although the 12-month group differences werenot significant forBMI orBMI percentile, the reduction in BMIdoubled from the 6-month assessment, and nearly doubled forBMI percentile among intervention participants. Controlparticipants at 6-months remained unchanged.

The evaluation data from parents and providers were fat/lean distribution over time may have confounded

bringing confidence to the validity of the tool. The evaluation

creases in intensity of counseling as child risks increase.

546 S.J. Tucker et al.overwhelmingly positive. Comments from parents andchildren reflect the strengths of the program including therelationship with the interventionist, inclusion of the entirefamily, the reinforcement in the follow-up calls, the problem-solving approach, and having someone else to work with thechild besides the parent working with the child. The findingsare consistent with those of Kubik, Story, Davey, Dudovitz,and Zuehlke (2008) who evaluated parent response to a clinic-based primary prevention intervention for increasing BMIscreening and counseling. In that study, significantly moreintervention parents reported intentions to ensure their childrenreceived five or more fruits/vegetables servings per day withinthe next 30 days. In the current study, providers likewisevalued the model and highlighted strengths of having thededicated nurse available to complement their care, wanting toenroll more of their children, having the resource available forfamilies, and the seamlessness in which the process worked.The study findings are also in line with outcomes by Schwartzet al. (2007) who examined the feasibility of primary carepediatricians and RDs implementing an office-based obesityprevention program using MI as the chief interventionstrategy. Like the current study, they found trends towardBMI percentile changes, positive parent evaluations andchallenges with retention of participants in the trial.

The study included a number of important methodologicallimitations that should be considered when interpreting results.First, there was no random assignment, thus other variablespresent in one group but not the other (e.g., parent obesityproblems, baseline levels of motivation for change) may haveinfluenced findings. The sample was fairly homogenous thuslimiting generalizability to families with the same demographicmake-up. The healthy habits data were all self-reported andparents and children may have over or underestimated theiractual habits, especially to look more favorable after theintervention period. Use of BMI as a primary endpoint has anumber of disadvantages including that (1) it is not an exactmeasure of body fat and highly muscular people may appearoverweight when they are not, (2) individuals who have lostweightmay have lost leanmassmore than fatmass and thus stillat risk despite a normal BMI, and (3) BMI does not distinguishabdominal fat, considered riskier for heart disease, from otherfat and, (4) BMI does not change easily. On the other hand,changes in weight are not possible to compare among childparticipants given the expected and variable changes in weightbased on child growth, and BMI is a convenient clinicalmeasure currently used in practice. Additionally, the statisticalsignificance was only a trend for BMI and as such may not be atrue or sustainable difference between groups. On the otherhand, insufficient power and missing data may have resulted ina type II statistical error leading us to falsely assume no effects.

The drop-out rate, common to intervention studies(Schwartz et al., 2007), is likewise acknowledged as limitingconclusions to those who remained in the study. Perhapsthose who needed the intervention the most were notreached. It is also possible that natural developmental andThus, all children are screened and offered basic 5-2-1-0healthy habits education, and as risks increase the nurseinterventionist becomes involved to offer MI. This modeladdresses concerns of providers' reports of not feelingcompetent in providing obesity care (Jay et al., 2008) orhaving the time to offer counseling (Passehl et al., 2004;Schwartz, 2010). It also introduces an MI intervention that isusable in real-world health settings, which Britt, Hudson, andBlampied (2004) identified as a remaining challenge for MI.The model has the potential to be cost-effective if it preventschildren from continued weight and related health problemsthat span into adulthood. It is also consistent with the goals ofthe health care home model and may prove an effectiveapproach that can be flexibly used at any patient visit wherethere is an opening to discuss weight concerns. The use ofRNs builds on the foundational tenets of nursing including astrong health promotion focus, relationship-based care, andintegration of environment into health interventions. Addi-tionally, strategies that maximize the RN role across settingsare consistent with health care reform actions and a goal ofthe Future of Nursing Report (Institute of Medicine, 2010).

Acknowledgments

Funding was provided by the Mayo Clinic PatientEducation Research Committee, Mayo Clinic NursingResearch Committee, and Small Grants Program sponsoredby the Mayo Clinic Center for Translational ScienceActivities. The authors acknowledge the financial supportand her encouragement of Dr. Jill Swanson, Consultant andChair of Community Pediatric & Adolescent Medicine.

References

Britt, E., Hudson, S. M., & Blampied, N. M. (2004). Motivational inter-viewing in health settings: A review. Patient Education and Counseling,53, 147155, http://dx.doi.org/10.1016/S0738-3991(03)00141-1.survey was simply a measure of satisfaction and parent/childfeedback about the quality of the program. These findings arebelieved to be valid perspectives of participants.

Despite these study limitations, the study findings haveimportant implications for clinical practice and supportfurther study. First, we tested a model that offers a teamapproach for screening with a RN to provide counselingusing MI for overweight children. The setting in which thestudy was conducted has since adapted this model for clinicalpractice with an algorithmic screening approach that in-findings. Last, the validity and reliability of the HealthyHabits Survey and the evaluation survey are not known andfindings are thus interpreted cautiously. That said, only itemlevel analysis was completed for the Healthy Habits Surveyand all items align with components of the 5-2-1-0 program

CDC (2011). CDC Grand Rounds: Childhood obesity in the United States.Morbidity and Mortality Weekly Report, 60(2), 4246.

Cowburn, G., Hillsdon, M., & Hankey, C. R. (1997). Obesity managementby life-style strategies. British Medical Bulletin, 53, 389408.

Daniels, S. R., Jacobson, M. S., McCrindle, B. W., Eckel, R. H., & Sanner,B. M. (2009). American Heart Association Childhood Obesity ResearchSummit Report. Circulation, 119, e489e517, http://dx.doi.org/10.1161/CIRCULATIONAHA.109.192216.

Dorsey, K. B., Wells, C., Krumholz, H. M., & Concato, J. (2005).Diagnosis, evaluation, and treatment of childhood obesity in pediatricpractice. Archives of Pediatrics & Adolescent Medicine, 159, 632638,http://dx.doi.org/10.1001/archpedi.159.7.632.

Institute of Medicine. (2010). The future of nursing: Leading change,

Ogden, C. L., Carroll, M. D., Curtin, L. R., Lamb, M. M., & Flegal, K. M.(2010). Prevalence of high body mass index in US children andadolescents, 20072008. Journal of the American Medical Association,303, 242249, http://dx.doi.org/10.1001/jama.2009.2012.

Oude Luttikhuis, H., Baur, L., Jansen, H., Shrewsbury, V. A., O'Malley, C.,Stolk, R. P., et al. (2009). Interventions for treating obesity inchildren. Cochrane Database of Systematic Reviews, CD001872,http://dx.doi.org/10.1002/14651858.CD001872.pub2.

Passehl, B., McCarroll, C., Buechner, J., Gearring, C., Smith, A. E., &Trowbridge, F. (2004). Preventing childhood obesity: Establishing healthylifestyle habits in the preschool years. Journal of Pediatric Health Care,18, 315319, http://dx.doi.org/10.1016/S0891524504002755.

Polacsek, M., Orr, J., Letourneau, L., Rogers, V., Holmberg, R., O'Rourke,K., et al. (2009). Impact of a primary care intervention on physician

547Reducing Pediatric Overweight(2008). Do internists, pediatricians, and psychiatrists feel competentin obesity care?: Using a needs assessment to drive curriculumdesign. Journal of General Internal Medicine, 23, 10661070,http://dx.doi.org/10.1007/s11606-008-0519-y.

Kamath, C. C., Vickers, K. S., Ehrlich, A., McGovern, L., Johnson, J.,Singhal, V., et al. (2008). Clinical review: Behavioral interventions toprevent childhood obesity: A systematic review and metaanalyses ofrandomized trials. Journal of Clinical Endocrinology and Metabolism,93, 46064615, http://dx.doi.org/10.1210/jc.2006-2411.

Krebs, N. S., & Jacobson, M. S. (2003, reaffirmed 2006). Prevention ofpediatric overweight and obesity. Pediatrics, 112, 424430.

Kubik, M. Y., Story, M., Davey, C., Dudovitz, B., & Zuehlke, E. U. (2008).Providing obesity prevention counseling to children during a primary careclinic visit: Results from a pilot study. Journal of the American DieteticAssociation, 108, 19021906, http://dx.doi.org/10.1016/j.jada.2008.08.017.

Let's Go! It's Working. Greater Portland Demonstration Project. RetrievedJune 1, 2012, from http://www.letsgo.org/wp-content/uploads/2011/11/LG5yrReport_Final.pdf.

Lobstein, T., Baur, L., & Uauy, R. (2004). Obesity in children and youngpeople: A crisis in public health. Obesity Reviews, 5(Suppl 1), 4104,http://dx.doi.org/10.1111/j.1467-789X.2004.00133.x.

Maine Center for Public Health. (2012) (2012). Keep me healthy. RetrievedFebruary 21, 2012, from http://www.mcph.org/major_activities/keepmehealthy.htm.

McAllister, E. J., Dhurandhar, N. V., Keith, S. W., Aronne, L. J., Barger, J.,Baskin, M., et al. (2009). Ten putative contributors to the obesityepidemic. Critical Reviews in Food Science and Nutrition, 49,868913, http://dx.doi.org/10.1080/10408390903372599.

McGovern, L., Johnson, J. N., Paulo, R., Hettinger, A., Singhal, V.,Kamath, C., et al. (2008). Clinical review: Treatment of pediatricobesity: a systematic review and meta-analysis of randomized trials.Journal of Clinical Endocrinology and Metabolism, 93, 46004605,http://dx.doi.org/10.1210/jc.2006-2409.

Miller, W. R. (1983). Motivational interviewing with problem drinkers.Behavioural Psychotherapy, 11, 147172, http://dx.doi.org/10.1017/S0141347300006583.

Miller, W. R., & Rollnick, S. (2002). Motivational interviewing: Preparingpeople for change (2nd ed.). New York: Guilford Press.practice and patient and family behavior: Keep ME HealthyTheMaine Youth Overweight Collaborative. Pediatrics, 123(Suppl 5),S258S266, http://dx.doi.org/10.1542/peds.2008-2780C.

Resnicow, K., Davis, R., & Rollnick, S. (2006). Motivational interview-ing for pediatric obesity: Conceptual issues and evidence review.Journal of the American Dietetic Association, 106, 20242033,http://dx.doi.org/10.1016/j.jada.2006.09.015.

Ritchie, L. D., Crawford, P. B., Hoelscher, D. M., & Sothern, M. S.(2006). Position of the American Dietetic Association: Individual-,family-, school-, and community-based interventions for pediatricoverweight. Journal of the American Dietetic Association, 106,925945.

Schwartz, R. P. (2010). Motivational interviewing (patient-centeredcounseling) to address childhood obesity. Pediatric Annals, 39,154158, http://dx.doi.org/10.3928/00904481-20100223-06.

Schwartz, R. P., Hamre, R., Dietz, W. H., Wasserman, R. C., Slora, E. J.,Myers, E. F., et al. (2007). Office-based motivational interviewing toprevent childhood obesity: A feasibility study. Archives of Pediatrics& Adolescent Medicine, 161, 495501, http://dx.doi.org/10.1001/archpedi.161.5.495.

Suarez, M., & Mullins, S. (2008). Motivational interviewing and pediatrichealth behavior interventions. Journal of Developmental andBehavioral Pediatrics, 29, 417428, http://dx.doi.org/10.1097/DBP.0b013e31818888b4.

Talmi, A., & Fazio, E. (2012). Commentary: Promoting health and well-being in pediatric primary care settings: Using health and behavior codesat routine well-child visits. Journal of Pediatric Psychology, 37,496502, http://dx.doi.org/10.1093/jpepsy/jss047.

Tucker, S., Lanningham-Foster, L., Murphy, J., Olsen, G., Orth, K.,Voss, J., et al. (2011). A school based community partnership forpromoting healthy habits for life. Journal of Community Health, 36,414422, http://dx.doi.org/10.1007/s10900-010-9323-9.

Waters, E., Swinburn, B., Seidell, J., & Uauy, R. (2010). Preventingchildhood obesity: Evidence policy and practice. Hoboken, NJ:Blackwell Publishing Ltd.

Whitlock, E. P., O'Connor, E. A., Williams, S. B., Beil, T. L., & Lutz, K. W.(2010). Effectiveness of weight management interventions in children:A targeted systematic review for the USPSTF. Pediatrics, 125,e396e418, http://dx.doi.org/10.1542/peds.2009-1955.advancing health. Washington, DC: Institute of Medicine.Jay, M., Gillespie, C., Ark, T., Richter, R., McMacken, M., Zabar, S., et al.

Reducing Pediatric Overweight: Nurse-Delivered Motivational Interviewing in Primary CareBackgroundMethodDesignSetting and SamplePower Estimate

Variables and MeasuresStandard Clinical CareMotivational Interviewing InterventionStudy ProceduresHuman Subjects ProtectionPhase 1 ProceduresPhase 2 Procedures

Data AnalysisResultsDemographic CharacteristicsStudy Aim 1: Compare Change in BMI Percentile and Self-Reported Health Habits of Children and Parents Between Those Who Rec.....Body Mass IndexHealthy HabitsIntent-to-Treat Analysis

2. Study Aim 2: Describe Parent, Child and Provider Satisfaction With the MI InterventionParentChild 6-Month EvaluationParentChild 12-Month EvaluationProvider 6-Month Evaluation

DiscussionAcknowledgmentsReferences