VOLUME 16 NUMBER 7 | JULY 2008 | www.obesityjournal.org
nature publishing group
A Pediatric Weight Management Program for
High-risk Populations: A Preliminary Analysis
Joseph A. Skelton1–3,8, Laure G. DeMattia4,5 and Glenn Flores6,7
Objective: To determine whether a multidisciplinary pediatric weight management program effectively improves BMI,
BMI z-score, and cardiovascular risk factors (CVRFs) in high-risk populations.
Methods and Procedures: A retrospective chart review was performed on children seen in the NEW Kids Program at
the Children’s Hospital of Wisconsin, a family-based clinic that treats pediatric obesity using medical management,
nutrition education, behavioral intervention, and physical activity. Inclusion criteria were program participation
for ≥9 months and >4 visits. Analyses were performed to identify factors associated with pre- to postintervention
changes in BMI, BMI z-score, and CVRF laboratory values.
Results: A total of 66 patients met inclusion criteria; the mean age was 11 years (s.d. ± 3.4), 56% were racial/ethnic
minorities, 45% were Medicaid recipients, 48% resided in impoverished communities, and 38% had a BMI
≥40 kg/m2. Of the 66 patients, 91% had more than one weight-related comorbidity, 88% had CVRFs, and the
preintervention mean BMI was 37 kg/m2. After the intervention, there was an overall increase in absolute BMI, but
a small, yet significant decrease in BMI z-score (mean –0.03 ± 0.16; P < 0.05). There were significant pregroup to
postgroup improvements in total cholesterol, low-density lipoprotein, and triglycerides levels (P < 0.05). Insurance
coverage, race/ethnicity, gender, age, and initial BMI were not significantly associated with changes in BMI
or BMI z-score.
Discussion: A multidisciplinary pediatric weight management program can improve the weight status of high-risk
populations, including minorities, Medicaid recipients, patients with multiple comorbidities and CVRFs, and the
Obesity (2008) 16, 1698–1701. doi:10.1038/oby.2008.243
Obesity has become pandemic in the United States, among
both children and adults (1). Unfortunately, few effective
treatment options exist for obese children. Interventions that
include behavioral change (2) and multidisciplinary treatment
teams (3) have yielded some success, but effective options for
childhood weight management are still limited, especially for
high-risk populations (1,4). Also, little is known about effec-
tive obesity treatments for children who are poor, have mul-
tiple comorbidities, or are morbidly obese (5).
The objectives of this study were to determine whether a
tertiary-care, multidisciplinary pediatric weight management
program (i) reaches high-risk populations, including those
who have multiple comborbidities, are poor, and are from
racial/ethnic minorities; and (ii) improves BMI, BMI z-score,
and cardiovascular risk factors (CVRFs).
ReseaRch Methods and PRoceduRes
the neW Kids Program
The NEW Kids Program at the Children’s Hospital of Wisconsin is a
multidisciplinary weight management clinic that features cognitive
behavioral modification in a team setting. Staffed jointly by a pediatri-
cian, nurse practitioner, dietitian, psychologist, and physical therapist,
this program is limited to 2- to 18-year-old children who are either
overweight or obese with one or more obesity-related comorbidities (6),
including dyslipidemia, insulin resistance, fatty liver, orthopedic prob-
lems, pseudotumor cerebri, sleep apnea, or hypertension.
Children are seen in individual clinic appointments by the multidisci-
plinary team; based on the initial visit, individualized care plans are for-
mulated for patients and families, with the entire household encouraged
to adopt healthy lifestyle changes. Therapy is based on a cognitive behav-
ioral model, in which the health psychologist works to increase patient
and family awareness of daily lifestyle habits, and transforms maladaptive
thoughts about lifestyle change into beneficial thoughts, feelings, and
1Division of Pediatric Gastroenterology and Nutrition, Department of Pediatrics, Medical College of Wisconsin, Milwaukee, Wisconsin, USA; 2NEW Kids Program,
Children’s Hospital of Wisconsin, Milwaukee, Wisconsin, USA; 3Children’s Research Institute, Children’s Hospital of Wisconsin, Milwaukee, Wisconsin, USA;
4Department of Family and Community Medicine, Medical College of Wisconsin, Milwaukee, Wisconsin, USA; 5DeMattia Medical Group, Milwaukee, Wisconsin, USA;
6Division of General Pediatrics, Department of Pediatrics, University of Texas Southwestern Medical Center, Dallas, Texas, USA; 7Children’s Medical Center, Dallas,
Texas, USA. 8Present address: Section of Pediatric Gastroenterology, Department of Pediatrics, Wake Forest University School of Medicine, Winston-Salem,
North Carolina, USA. Correspondence: Joseph A. Skelton (email@example.com)
Received 21 February 2007; accepted 1 October 2007; published online 24 April 2008. doi:10.1038/oby.2008.243
Obesity | VOLUME 16 NUMBER 7 | JULY 2008 1699
behaviors. Nutrition and physical activity education, training, and sup-
port are provided to assist caregivers in implementing behavioral change.
Problem areas in nutrition and activity are identified and prioritized, with
specific recommendations focused on deficient areas. Advancement to
more rigorous dietary and exercise treatment (portion control, increased
complex carbohydrates, exercise prescriptions) is attempted after progress
has been made in addressing the problem areas. Most families make
visits every 1–3 months. As patients and families demonstrate success
(i.e., positive social, emotional, and/or behavioral change, improvements
in laboratory studies, and/or improved BMI), appointment intervals are
increased until patients no longer meet program criteria, resulting in
referral back to their primary care physician.
study design and chart abstraction
A retrospective chart review was performed by program physicians
and trained research assistants on patients initially evaluated in the
NEW Kids Program from 1 June 2003 to 1 April 2005. Inclusion cri-
teria were program participation for ≥9 months, and more than four
visits to the clinic; these criteria were chosen to examine specifically
the program’s effect on patients who adhered to follow-up recommen-
dations and had adequate exposure to the clinic and its interventions.
It typically takes >4 visits to complete basic nutrition education and
establish behavioral change patterns, with treatment duration typi-
cally ~12 months.
Sociodemographic data collected during program enrollment (base-
line) included birth date, self-reported parental height and weight, zip
code, race/ethnicity, and guardianship. Additional baseline data col-
lected included initial fasting serum laboratory studies (insulin, glu-
cose, hemoglobin A1C, aspartate transaminase, alanine transaminase,
total cholesterol, low-density lipoprotein, high-density lipoprotein, and
triglycerides) from the referring physician, and also height and weight.
Weight-related comorbidities were diagnosed by the NEW Kids cli-
nicians through review of laboratory studies, initial evaluation, past
medical history, and medical records. Laboratory studies and anthro-
pometric data were collected at each follow-up visit. All height and
weight measurements were performed by trained NEW Kids team
members using a Harpenden stadiometer (Holtain, Dyfed, UK) and
Health o meter Pro Plus electronic wheelchair scale (Pelstar, Alsip, IL),
with the child’s shoes removed and in light clothing.
Wilcoxon test and ANOVA were used to analyze significant changes in
BMI, BMI z-score, and laboratory values before and >9 months post-
program initiation. χ2- and paired t-tests were used to examine asso-
ciations between sociodemographic factors and changes in BMI, BMI
z-score, and laboratory values. All statistical analyses were performed
using SAS version 8.2 (SAS Institute, Cary, NC).
study sample and participant characteristics
Initial evaluations of 398 patients were done between June 2003,
and April 2005; 23 patients were excluded who were seen for a
single-visit consultation to evaluate endogenous obesity causes,
127 patients were excluded because of program enrollment for
<9 months, and 182 dropped out of the program (Table 1).
The remaining 66 patients who met the study inclusion cri-
teria made a mean (±s.d.) of 10 (±4.1) visits to the program
over a mean follow-up duration of 13 (±3) months. A total of
48% children resided in impoverished areas of the central city
(defined as residing in zip codes within Milwaukee City limits
and designated as urban and with the highest rates of families
living in poverty, according to the US Census Bureau, http://
Baseline BMI, attrition, and comorbidities
Study children were substantially obese (Table 1), with a mean
(±s.d.) BMI at the 99th (±6) percentile, with 38% having a
Limited sociodemographic, height, and weight data were col-
lected on the 182 children (73%) dropping out of the program
during the study period (dropouts) (Table 1). Sociodemographic
characteristics of the 182 dropouts did not significantly dif-
fer from those who met the study criteria. Attrition rates were
similar between groups; whether examined by race/ethnicity
(white 73%, minority 74%), insurance coverage (commercial
77%, Medicaid 67%), gender (female 76%, male 70%), and
caregiver category (dual parent 75%, single 76%).
The vast majority of study children (91%) had two or more
overweight-related comorbidities (Table 2). Two-thirds of
children had elevated fasting insulin levels (Table 3). Patients
were taking a median of two prescription medications.
Pre- to postintervention changes in study outcomes
There was no pre- to postintervention change in the median
BMI, and a statistically significantly increase in mean BMI.
There was a significant decrease in BMI z-score (Table 4).
Significant pre- to postintervention reductions occurred in
three of five CVRFs (Table 5). Outcomes did not significantly
differ between more and less severely obese patients, and both
groups experienced reductions in CVRFs (data not shown).
table 1 selected baseline data for study participants:
sociodemographic characteristics and BMI data for
participants in the neW Kids Program
Mean age in years (±s.d.) 11.8 (±3.4)11.3 (±3.8)
Male gender50% 42%
Mean BMI, kg/m2 (±s.d.) 37 (±10) 35 (±8.8)
African American 39%35%
Health insurance coverage
Single parent32% 38%
HMO, Health Maintenance Organization.
aNo statistically significant differences between study children and dropouts.
VOLUME 16 NUMBER 7 | JULY 2008 | www.obesityjournal.org
Insurance coverage, race/ethnicity, gender, age, and initial
BMI were not significantly associated with either BMI or BMI
The findings of this preliminary analysis suggest that a mul-
tidisciplinary pediatric weight management program can
improve weight status and CVRFs in a diverse population. The
study intervention was as effective for racial/ethnic minor-
ity children as it was for white children in reducing BMI and
CVRFs. This finding is noteworthy given that minority chil-
dren continue to have the highest obesity rates (1,4,7). Several
features might account for the success of the study program
among minority children and families, including approaches
tailored to each family, and emphasis on culturally appropriate
education and information.
The most successful pediatric weight management inter-
ventions have been evaluated in predominantly dual parent,
middle-class white populations with few comorbidities (2), but
few studies exist for diverse populations. One study reported
successful outcomes among children whose mean BMI was
35.6 kg/m2, with 32% African American and 32% receiving
Medicaid (8). A recent randomized, controlled study demon-
strated improvement in weight status and insulin resistance,
with over half of participants being racial/ethnic minorities (9).
Our study had a high attrition rate (73%) across all groups.
A high rate (83%) was noted initially in the recent controlled
study (9), falling at the upper range of reported rates in pedi-
atric programs (33–83%) (8–13). Programs have begun to
explore attrition determinants (12,13), as this is an important
part of behavioral medicine, and could improve overall treat-
Certain study limitations should be noted. Although there
were several statistically significant findings, the sample size
and overall change in BMI of this preliminary study was rela-
tively small. The statistical power of this study was insufficient
to determine whether change in BMI z-score was associated
with changes in laboratory values. In-depth analyses could
not be performed for subjects who withdrew or were lost to
follow-up, so the findings may have been subject to potential
distortion due to lack of data on these subjects, although analy-
ses revealed no differences between participants and dropouts.
Because this preliminary analysis evaluated a relatively small
sample size of patients, these findings may not necessarily gen-
eralize to other populations, and randomized, controlled trials
of the program effectiveness are needed to confirm the find-
ings of this retrospective study without a control group.
Although the statistically significant changes in BMI and
BMI z-score were not substantial, we believe that they are
of clinical significance, as the growth trajectory of children
enrolled in the program was quite steep, often outstripping
the growth curve. Given that most obese children are likely to
become overweight or obese adults (14), programs are needed
table 3 selected baseline data for study participants:
baseline laboratory values for study children (N = 66) prior to
participation in the neW Kids program
Insulin (U/l) 56 30 (±30)
Triglycerides (mg/dl)62 134 (±76) 52
LDL (mg/dl)59 111 (±31)47
HDL (mg/dl) 6541 (±11) 29
≤170 mg/dl 41
HDL, high-density lipoprotein; LDL, low-density lipoprotein.
aSome patients did not have all laboratory studies performed, due to incomplete
records from referring physicians or missed opportunities by staff.
table 4 Pre- to postintervention changes in BMI and BMI
z-scores among study children (N = 66) in the neW Kids
Weight status change
Overall BMI change1.0 kg/m2 (±2.7)*0 kg/m2 (–6,11)
Overall BMI z-score change–0.03 (±0.16)** –0.03 (–0.7, 0.3)**
*P = 0.005 by Wilcoxon test, **P < 0.05 by Wilcoxon test.
table 5 Pre- to postintervention changes in laboratory values
among study children (N = 66) in the neW Kids Program
Total cholesterol58 –11.7 (±22)* –12.5 (–69, 36)*
Triglycerides 57–10.3 (±65)**–15 (–251, 176)**
LDL 55 –7.1 (±25)*** –6 (–55, 78)***
HDL 610.3 (±11)†
3 (–28, 28)†
Insulin 49 –6.5 (±21)†
–2 (–90, 24)†
HDL, high-density lipoprotein; LDL, low-density lipoprotein.
*P < 0.001, **P = 0.05, ***P < 0.05, †not statistically significant.
table 2 selected baseline data for study participants:
prevalence of obesity-related comorbidity among study
children (N = 66) in the neW Kids Program
Proportion of children with
Elevated liver function test12
Two or more comorbidities91
aDefined as abnormal fasting level of low-density lipoprotein, total cholesterol, or
triglycerides. bDefined as consistent shortness of breath or inability to keep up
with peers in physical activity, by patient self-report or parent proxy-report.
Obesity | VOLUME 16 NUMBER 7 | JULY 2008 1701 Download full-text
3. Sothern MS, von Almen TK, Schumacher H et al. An effective
multidisciplinary approach to weight reduction in youth. Ann NY Acad Sci
4. Ogden CL, Flegal KM, Carroll MD, Johnson CL. Prevalence and trends
in overweight among US children and adolescents, 1999–2000. JAMA
5. Gordon-Larsen P, Adair LS, Nelson MC, Popkin BM. Five-year obesity
incidence in the transition period between adolescence and adulthood:
the National Longitudinal Study of Adolescent Health. Am J Clin Nutr
6. Dietz WH, Robinson TN. Clinical practice. Overweight children and
adolescents. N Engl J Med 2005;352:2100–2109.
7. Kimm SY, Barton BA, Obarzanek E et al. Obesity development during
adolescence in a biracial cohort: the NHLBI Growth and Health Study.
8. Kirk S, Zeller M, Claytor R et al. The relationship of health outcomes to
improvement in BMI in children and adolescents. Obes Res 2005;13:
9. Savoye M, Shaw M, Dziura J et al. Effects of a weight management program
on body composition and metabolic parameters in overweight children:
a randomized controlled trial. JAMA 2007;297:2697–2704.
10. Tershakovec AM, Kuppler K. Ethinicity, insurance type, and
follow-up in a pediatric weight management program. Obes Res 2003;11:
11. Levine MD, Ringham RM, Kalarchian MA, Wisniewski L, Marcus MD. Is
family-based behavioral weight control appropriate for severe pediatric
obesity? Int J Eat Disord 2001;30:318–328.
12. Barlow SE, Ohlemeyer CL. Parent reasons for nonreturn to a
pediatric weight management program. Clin Pediatr (Phila) 2006;45:
13. Cote MP, Byczkowski, Kotagal U et al. Service quality and attrition:
an examination of a pediatric obesity program. Int J Qual Health Care
14. Guo SS, Wu W, Chumlea WC, Roche AF. Predicting overweight and obesity
in adulthood from body mass index values in childhood and adolescence.
Am J Clin Nutr 2002;76:653–658.
that are efficacious in reducing BMI, BMI z-scores, and CVRFs,
particularly among hardest-to-reach populations and greatest
In summary, a multidisciplinary pediatric weight manage-
ment program can achieve success in high-risk populations,
including racial/ethnic minorities, the poor, the severely obese,
and patients with multiple comorbidities and CVRFs. The
findings of this preliminary analysis suggest that a multidis-
ciplinary pediatric weight management program could be an
effective means of reaching and treating high-risk populations
of obese children.
We thank Maureen Otto, Qun Xiang, Jing-nan Mao, George Kay and the
Division of Quantitative Health Sciences, Department of Pediatrics, Medical
College of Wisconsin for their statistical and organizational help in this
project. This study was presented in part as a poster at the annual meetings
of the Pediatric Academic Societies, 30 April 2006, in San Francisco, CA,
and the North American Association for the Study of Obesity, 18 October
2005, in Vancouver, British Columbia, Canada.
The authors declared no conflict of interest.
© 2008 The Obesity Society
1. Hedley AA, Ogden CL, Johnson CL et al. Prevalence of overweight and
obesity among US children, adolescents, and adults, 1999–2002. JAMA
2. Epstein LH, Valoski A, Wing RR, McCurley J. Ten-year outcomes of
behavioral family-based treatment for childhood obesity. Health Psychol