Changing Adolescent Health Behaviors

The Healthy Teens Counseling Approach

Ardis L. Olson, MD, Cecelia A. Gaffney, MEd, Pamela W. Lee, PhD, Pamela Starr, MS

Background: Brief motivational interventions that have been provided in addition to routine primary
care have changed adolescent health behaviors. Whether health screening and motivational-
interviewing– based counseling provided by clinicians during routine care can change
behaviors is unknown.
Methods: Healthy Teens was a primary care, office-system intervention to support efficient, patient-
centered counseling at well visits. Healthy Teens utilized a personal digital assistant
(PDA)– based screener that provided the clinician with information about a teen’s health
risks and motivation to change. Changes in adolescent self-report of diet and activity health
behaviors 6 months later were assessed in two cross-sectional samples of teens from five
rural practices in 2005 and 2006. Usual-care subjects (N⫽148) were recruited at well visits
prior to the intervention, and the Healthy Teens subjects (N⫽136) were recruited at well
visits after the Healthy Teens system was well established.
Results: At 6-month follow-up, the Healthy Teens group had significantly increased self-reported
exercise levels and milk-product intake. In the models exploring covariates, the only
significant predictors for improvement in exercise levels were intervention-group status
(p⫽0.009) and post-visit interest in making a change (p⫽0.015). Interest in changing
predicted increased milk intake (p⫽0.028) in both groups. When teens planned an action
related to nutrition, physical activity, or both after a well visit, Healthy Teens participants
were more likely to report multiple planned actions (68% Healthy Teens vs 32% usual care,
p⬍0.05).
Conclusions: Changes in office systems using low-cost technology to screen adolescents and promote
patient-centered counseling appear to influence teens to increase exercise and milk intake.
(Am J Prev Med 2008;35(5S):S359 –S364) © 2008 American Journal of Preventive Medicine

Introduction be honest about sensitive issues like substance use.7 But

screening-questionnaire use has not been widely em-

A
dolescent health– compromising behaviors can
braced by clinicians, and oral questioning has been
persist into adulthood and contribute to
inconsistent. For example, only 5% of pediatricians
chronic disease and mortality. National surveys
reported routinely using screening questionnaires that
reveal that 70% of adolescents report one or more of
eight health-risk behaviors.1 To address these chal- assessed substance use.8
lenges, national guidelines recommend screening and The Healthy Teens intervention, designed to en-
preventive services for adolescents.2,3 Adolescents con- hance the adolescent well visit, included (1) a compre-
sider healthcare providers a credible source of infor- hensive health- and behavior-risk screener via a low-cost
mation, and most want to discuss health risks with their personal digital assistant (PDA (e.g., Palm®); (2) clini-
clinician.4 While most adolescents apparently want to cian training in brief motivational-interviewing tech-
discuss risk behaviors, clinician inquiry and discussion, niques, complemented by information from the PDA
however, are infrequent.5,6 When adolescents respond screening that prompted clinicians to use a motivational-
to computer or paper screening, they are more likely to interviewing approach; and (3) information about out-
side resources for practices and adolescents. The hy-
pothesis was that adolescents who received well visits
From the Department of Pediatrics (Olson) and the Department of enhanced by the Healthy Teens intervention would
Community and Family Medicine (Olson, Gaffney, Lee, Starr),
Dartmouth Medical School, Lebanon, New Hampshire
later report improved health behaviors compared to
Address correspondence and reprint requests to: Ardis L. Olson, adolescents seen in these practices prior to the Healthy
MD, Professor of Pediatrics and Community and Family Medicine, Teens implementation. The Healthy Teens project was
Clinicians Enhancing Child Health Network, Dartmouth Hitchcock
Medical Center, Department of Pediatrics HB 7450, Lebanon NH supported by the second round of the Robert Wood
03756-001. E-mail: [email protected]. Johnson Foundation Prescription for Health program.

Am J Prev Med 2008;35(5S) 0749-3797/08/$–see front matter S359

© 2008 American Journal of Preventive Medicine • Published by Elsevier Inc. doi:10.1016/j.amepre.2008.08.014
Methods teen’s interest in making a change and perceived importance
and confidence that the teen could change each specific
Study Design, Context, and Subjects behavior. Prior to the visit, the clinician reviewed a PDA
The Healthy Teens project was conducted in five rural summary of the patient’s health concerns, risky and healthy
primary care practices using a pre–post evaluation study behaviors, and interest in change. All clinical staff and
design from 2005 to 2006. Prior to the Healthy Teens clinicians received the same training in using the PDA, and
implementation, a control group of adolescents with usual received assistance in incorporating this technology into
care was enrolled after well visits. Clinicians and practices existing office operations.
were then trained to implement the Healthy Teens interven- Clinician training in brief motivational-interviewing coun-
tion. After Healthy Teens components had been in regular seling skills was provided to support patient-centered coun-
use for at least 6 months, a second evaluation group of seling because of its potential to improve health behav-
adolescents who received the Healthy Teens intervention was iors.12–14 All clinicians received 3 hours of interactive training
enrolled after well visits. that included reflective listening, methods of addressing
Practices in New Hampshire and Vermont from the ambivalence, and goal setting. Training was provided at all
practice-based research network Clinicians Enhancing Child sites by a clinician and health psychologist.
Health were selected to include a range of practice size (1–9 Each site had three lunchtime presentations by community
full-time equivalents) and specialty mix (family medicine [1]; services to enhance links to local resources. Practices chose
pediatric [3]; and mixed family medicine/pediatric [1]). In presentations from available programs based on interest, so
these practices, patients were 95% Caucasian, and had Med- topics and content varied across sites. Presenters ranged from
icaid insurance rates varying from 10% to 40%. police departments to mental health services to teen activity
The evaluation groups were recruited consecutively at teen programs. Because of limited local resources for supporting
health visits over two 3-week periods in July–August 2005 and behavior change, a pocket card of regional and national
July–August 2006. First, teens completed in the office an web-based resources and toll-free numbers was developed.
anonymous exit survey at baseline. Those who consented to Measures
enrollment were mailed a follow-up survey 6 months later.
The intervention sample was recruited 1 year later to have The evaluation questions were selected by the Prescription
similar seasonal variations in both samples. Of adolescents in for Health program office10; they measured current adoles-
the usual-care group who completed the baseline survey, 76% cent health behaviors regarding eating patterns (consump-
enrolled in the study. Of the potential intervention group, tion of fruits and vegetables, milk, and sweetened beverages);
87% enrolled. Small financial incentives were given for the physical activity excluding physical education (days/week
initial ($3) and follow-up survey ($10). Adolescents aged ⱖ18 when moderately active for 30 minutes or more, hours of
years signed informed consent at enrollment, and those aged weekday television and non-academic computer use); ciga-
⬍18 years provided assent along with parental consent. The rette smoking; and alcohol consumption.
Dartmouth Medical School IRB approved the study protocol. In the initial survey, for each of the above-mentioned
Clinicians were surveyed both prior to training and 18 behaviors, subjects were asked if the topic had been discussed
months later to assess changes in their attitudes and per- in the visit. If the topic had been discussed, the subject’s
ceived counseling skills as well as in their views about PDA interest in making a change was determined. Response op-
use. The post-survey was taken during the maintenance tions were a simplification of the stages-of-change model15
period when all practices had used the Healthy Teens system and included doing well no need, not interested, thinking about
for at least 12 months. changing, willing to try to change, and already trying to change.
The Healthy Teens evaluation was guided by the Reach, Subjects with any active interest (thinking . . ., willing to try . . .,
Efficacy/Effectiveness, Adoption, Implementation, Mainte- or already trying to change) were considered to be interested in
nance (RE-AIM) framework.9,10 The adolescent-outcome change after clinician interaction. Those who responded no
measures addressed the efficacy of the intervention. Reach need or not interested, or who never discussed the topic, were
was measured by the percentage of teens that completed PDA classified as not interested in change.
screeners during health visits at 18 months. The uptake of the Subjects in the initial survey were also asked to list any
program by practices was a measure of adoption, and main- change they planned to make following this visit with their cli-
tenance was based on the number of clinicians who used nician. Responses were coded as a planned nutrition change, a
PDAs during the project and then reported the intention to planned physical activity/sedentary behavior change, or some
continue PDA use at 18 months. other planned health behavior change (e.g., alcohol/drugs,
sexual health, mental health). Subjects with any response
were classified as having made an action plan for that item.
Intervention
The total number of planned actions was calculated for each
The intervention provided clinicians with tools and strategies subject.
to incorporate into the well visit. Teens completed a screener A clinician survey assessed perceived counseling skills and
via PDA in the office prior to clinician contact. Screener roles prior to and 18 months after the Healthy Teens imple-
questions were derived from existing adolescent health be- mentation. Clinicians were asked their level of agreement
havior screening questionnaires as well as from the sugges- with statements using a 5-point scale (1⫽strongly agree to
tions of a panel of practicing pediatricians and family physi- 5⫽strongly disagree). Post-surveys included additional items
cians.11 For the health behaviors targeted by the Prescription pertaining to motivational-interviewing skills and PDA imple-
for Health program (tobacco use, unhealthy diet, physical mentation. Perceived confidence in these skills and the use of
inactivity, and risky alcohol use), the screener assessed the a PDA were assessed post-intervention only.

S360 American Journal of Preventive Medicine, Volume 35, Number 5S www.ajpm-online.net

Data Analysis Results
The current analysis focused on comparing the self-reported Delivery of the Intervention
changes in health behaviors after 6 months between the two Among clinicians, 11% (4/35) chose not to participate
teen groups. To assess the full impact of the Healthy Teens
and did not use PDAs. These nonparticipating clini-
program, the evaluation samples were limited to 136 teens in
cians were similar to participating clinicians in age and
the usual-care group who had not used the PDA screener and
to 148 teens in the Healthy Teens group who had PDA gender. Adolescent health risk screening via PDA was
screening. Baseline comparisons of age, gender, and all established as a routine in the five practices. Over 15
health-risk behaviors between the analysis group and the months, PDA screening was completed by 1670 teens.
subjects excluded by the above criteria showed no significant The PDA was used in 68%–74% of all adolescent health
differences using t-tests and chi-square. The same statistical visits. Office staff said that patient refusal of PDA
method was used to compare the baseline characteristics of screening seldom occurred. The most common re-
the final usual-care and Healthy Teens groups and to deter- ported reason for non-use were breaks in office screen-
mine whether health risks at baseline differed for completers ing routines and late arrival.
and noncompleters of the follow-up survey. Of the 31 clinicians who participated in the study,
Health behavior change scores were calculated for each 23 completed the initial survey, 24 completed the
subject’s Prescription for Health health behavior by subtract-
follow-up survey, and 16 completed both. After the
ing baseline from 6-month responses. The sign was reversed
for screen time and sweetened beverages so that a positive
intervention, there were two changes in the surveyed
change was always an improved health behavior. clinicians’ perceptions of their health counseling.
The primary outcomes analyses assessed average changes in While their views of counseling roles and effectiveness
health behaviors after 6 months and the predictors of were unchanged, the clinicians found health counsel-
changes in health behaviors. First, t-tests were performed to ing easier and thought that they listened better (Table
examine group differences. Next, univariate ANCOVA anal- 1). The majority of all providers perceived that use of
yses, controlling for practice site, were performed for fruits the PDA enhanced their visit and expressed confidence
and vegetables, milk, sweetened beverages, physical activity, in new motivational-interviewing skills 18 months after
and screen time. Covariates were intervention status, gender, training (Table 2). In addition, 75% of the post-survey
discussion of the topic in the visit, and interest in changing respondents reported that they definitely planned to
the behavior. Age showed no differences in univariate analy-
continue using the PDAs.
ses and was excluded. Because of the wide age range, two of
the health risks (recent alcohol or tobacco use) had low
prevalence, and the number of subjects was too small to Adolescent Outcomes
analyze. Only variables with complete data in analyses were Surveys at the 6-month follow-up were completed by
used, with no imputations for missing data. 68% of subjects in both the usual-care and Healthy
Ongoing PDA usage was tracked by the monthly download-
Teens group (usual care, 92/136; Healthy Teens, 101/
ing of Health Insurance Portability and Accountability Act of
1996 (HIPAA)– delimited data from PDA screeners. Appoint-
148). Respondents were aged 11–20 years. Age, gender,
ment records of the number of health visits and PDA screener and ethnicity did not differ significantly by group
data for 1 month were compared at 15 months’ post-training. (Table 3). Baseline health behaviors differed between
Clinician surveys were analyzed with paired t-tests to compare groups only for sweetened-beverage intake and days/
means of pre–post items. All analyses were conducted using week with moderate physical activity (Table 2). Baseline
SPSS version 13. characteristics were not different between subjects in

Table 1. Clinician self-perceived counseling skills and responsibilities, prior to and after Healthy Teens intervention (N⫽16)
Survey item Pre-PDA use Ma (SD) With PDA use Ma (SD) p-value
In general, it is easy to incorporate health 2.94 (0.85) 2.31 (0.70) 0.01
behavior counseling in my daily practice.
I am a good listener with my patients. 2.25 (0.68) 1.88 (0.62) 0.03
I am effective in helping patients change. 2.56 (0.63) 2.50 (0.63) 0.58
I do not have enough time to counsel patients 2.88 (1.09) 2.81 (1.17) 0.86
about changing health behaviors.
It is important for me to counsel my patients 1.25 (0.45) 1.25 (0.45) 1.00
about changing health behaviors.
I need to learn new strategies for helping my 1.67 (0.82) 2.07 (0.70) 0.11
patients change health behaviors.
It is my responsibility to determine the 2.63 (1.20) 2.25 (0.86) 0.30
patient’s priorities for the visit.
Note: Analysis by paired t-test. Boldface text designates which findings were significant at p⬍0.05 level.
a
1–5 scale from strongly agree to strongly disagree
PDA, personal digital assistant

November 2008 Am J Prev Med 2008;35(5S) S361

Table 2. Post-intervention clinician self-perceptions of 15% who reported alcohol use in the past month, if
motivational-interviewing skills and use of the personal clinicians discussed the topic, only 15% (6/37) left the
digital assistant (PDA) screener (N⫽24) visit interested in making a change in their drinking. In
Agree/strongly contrast, among teens with low exercise levels (⬍3days/
disgreea week), 90% of the Healthy Teens and 65% of the
Survey item n (%) usual-care teens left the visit interested in making a
I feel confident: change to be more active after discussion with the
using reflective listening 17 (70.8) clinician.
discussing pros and cons 18 (75.0) Comparisons of unadjusted average changes in teen
using importance/confidence scaling 10 (43.5)
questions health behaviors from baseline to 6 months between
Using the PDA screener: Healthy Teens and usual-care teens are provided in
helps me identify sensitive issues 21 (87.5) Table 4. All changes in behaviors favored the interven-
helps me set priorities for the visit 20 (83.3) tion group except for those involving sweetened bever-
allows me to use the time in the visit 18 (75.0)
ages. However, the only significant changes were for
more effectively
a
milk intake and physical activity.
1–5 scale from strongly agree to strongly disagree
Controlling for practice site, ANCOVA models that
predicted health behavior change over 6 months were
Healthy Teens or usual care among completers of significant for both milk intake and physical activity
6-month surveys and those lost to follow-up, except that (both p⫽0.01). Specific predictors of improvement in
usual-care noncompleters were more likely to drink ⬎1 physical activity level after 6 months were the Healthy
sweetened beverage per day (83% vs 64%, p⫽0.04). Teens intervention group (p⫽0.009) and an interest in
After the visit, there was a trend for teens in the making a change at baseline (p⫽0.015). The interac-
Healthy Teens intervention group to report a speci- tion of the Healthy Teens group with interest in making
fic nutrition/physical activity planned action (42% a change played a lesser role (p⫽0.09). Gender did not
Healthy Teens vs 32% usual care, p⫽0.097). If they emerge as a significant predictor in the model
planned specific actions, the Healthy Teens cohort was (p⫽0.447), nor was there interaction between gender
significantly more likely to have reported multiple and intervention status. The only significant predictor
intended actions (⬎1 planned action, 68% Healthy of improved milk intake was interest in making a
Teens vs 32% usual care, p⬍0.05). change at baseline (p⫽0.028).
While the number of adolescents in these samples
currently using substances (tobacco, alcohol) was too
Discussion
small to analyze outcomes, preliminary data were ob-
tained about the challenge for clinicians in motivating This study was designed to assess the potential effective-
adolescents with recent use to change their consump- ness of coupling PDAs for health behavior assessment
tion. Among the 11% who had used tobacco in the with training and prompts that supported the use of
prior month, 35% (9/26) left the visit interested in brief motivational-interviewing techniques to counsel
making a change after clinician discussion. Among the about health risks. Adolescents participating in this

Table 3. Participant characteristics and self-reported health behaviors at baseline

Intervention (nⴝ148) Usual care (nⴝ136) p-value
Characteristics (%)
Female 50.0 47.3 0.37
White 96.0 93.5 0.32
Hispanic ethnicity 3.5 2.2 0.47
Aged ⱕ14 years 45.3 44.9 0.52
Baseline health behavior (M, SD)
Screen timea (hours/weekday)b 3.89 (3.80) 3.93 (3.49) 0.93
Physically active ⱖ30 minutes (days/last week) 4.25 (2.09) 4.86 (1.83) 0.01
Sweetened beverages (servings/days)b 2.45 (2.54) 3.36 (3.12) 0.01
Glasses of milk (8 oz servings/day)b 2.28 (1.74) 2.27 (1.64) 0.95
Fruits and vegetables (servings/day)b 3.69 (1.93) 3.91 (2.21) 0.38
Used alcohol in the past month (%) 11.8 16.3 0.27
Among drinkers, days with ⱖ1 alcoholic drink 0.45 (1.38) 0.69 (2.67) 0.35
Smoked in past month (%) 10.1 8.8 0.71
Among smokers, days smoked 6.67 (11.49) 6.43 (10.20) 0.93
Note: Boldface text designates which findings were significant at p⬍0.05 level.
a
Screen time includes television and computer games.
b
Item asks about health behavior for a typical day.

S362 American Journal of Preventive Medicine, Volume 35, Number 5S www.ajpm-online.net

Table 4. Average changea in health behaviors from baseline to 6 months
Change in health behaviorsb Intervention (nⴝ101) Usual care (nⴝ92) p-value
Physical activity ⱖ30 minutes (days/week) 0.581 ⫺0.220 0.006
Fruits and vegetables (servings/day) 0.165 ⫺0.094 0.386
Milk (servings/day) 0.190 ⫺0.313 0.012
Sweetened beverages (servings/day) ⫺0.151 0.638 0.059
Screen timec (hours/weekday) 0.687 0.286 0.414
Note: Boldface text designates which findings were significant at p⬍0.05 level.
a
t-test (⬍5 missing subjects for any individual health behavior)
b
Self-reported
c
Screen time includes television and computer games.

primary care intervention utilizing low-cost technology community linkages. It is recognized that the adoles-
were more likely to be interested in addressing some cents were not randomly selected and depended on
obesity-related health behaviors and to plan specific their families’ willingness to stay after the visit to
behavior-change actions. Six months later, adolescents complete the initial survey. Adolescents who attend well
who had a Healthy Teens visit reported significant visits have fewer health risks than adolescents screened
increases in their amount of exercise. Effective screen- in schools.19 Given these factors, along with subjects
ing and skill-building among clinicians supported more lost to follow-up, it is possible that a population may
patient-centered counseling that may have activated have been selected that was more amenable to address-
patients. These results are consistent with those of ing some issues. The limited number of practices and
Patrick et al.,16 who found that a stage-of-change– based small samples allowed control at the practice level but
approach, using interactive computer programs at teen not at the clinician level. Only 52% of clinicians re-
health visits with additional health counselors, im- sponded to pre–post surveys, but similar positive clini-
proved diet and physical activity but had a greater effect cians’ appraisals of the program’s impact were found in
when the teen set a goal of improving physical activity. follow-up surveys completed by 77% of the clinicians.
A similar program has been effective for smoking These study results are promising in several ways. In
cessation, but did not increase fruit and vegetable contrast to other primary care interventions that tai-
intake.17 lored counseling for health-risk behaviors, the Healthy
Why were no improvements found in nutrition out- Teens intervention was incorporated into existing prac-
comes? The nutrition measures selected by the pro- tices without requiring additional staff or computer
gram office assessed the typical-day intake of several access for patients.16,17 Teens used PDAs while waiting
items rather than the 24-hour recall or 3-day diaries to be seen, reducing staff and clinician time to gather
used in other adolescent studies that have improved and review health information. This allowed the clini-
intake.16,18 The less-precise recall measures of dietary cian to set priorities and use the limited time for
intake used in this study may obscure dietary changes. counseling. This appears to have been most effective
The limited nutrition items also did not capture for exercise. While beyond the scope of this report, the
broader changes in nutrition planned by the subjects Healthy Teens approach helped clinicians discuss sen-
(e.g., stop eating so much food, make healthier choices). The sitive topics such as family and peer concerns as well as
PDA assessed teen interest in “eating healthier,” not in emotional and sexual issues. The provision of compre-
changing any of the specific survey items. This less- hensive screening, along with more in-depth informa-
specific interest measure may have triggered nutrition tion if risks existed, may explain why clinicians planned
discussions not captured here. The focus of the Healthy to keep using the PDAs after the study.
Teens program on healthy behaviors, not weight loss, In summary, a review of the Healthy Teens program
also may have contributed to fewer reported changes in within the RE-AIM framework found that the impact of
food intake. Unfortunately, there are no height or the program was positive for each component. Nearly
weight data on this population to determine the out- three quarters of the patients in participating practices
comes for overweight adolescents. were reached. Adolescents who were screened and
There are additional limitations to consider in inter- received enhanced counseling reported improved
preting these findings. This is a small study that ex- physical activity (efficacy). The Healthy Teens program
plored whether modest changes in clinician screening was flexible at the practice level and allowed clinicians
and brief training can influence adolescents’ health to use the tools differently, leading to adoption by most
behaviors. Limited resources and the desire to imple- clinicians. Project staff facilitated uptake but implemen-
ment an office-level intervention that could be realisti- tation efforts were local, leading to implementation
cally disseminated did not allow extensive time for across all sites. The computerized screening ensured
training in motivational interviewing, direct assessment the fidelity of that component, and clinicians’ self-
of clinicians’ delivery of counseling, or more extensive report of enhanced counseling skills indicated that

November 2008 Am J Prev Med 2008;35(5S) S363

their training was helpful. The fact that Healthy Teens 2. American Academy of Pediatrics Committee on Practice and Ambulatory
Medicine. Recommendations for preventative pediatric health care. Pedi-
program components were integrated into existing atrics 2000;105:645– 6.
office systems utilizing current staff led to maintenance 3. Elster A. The American Medical Association guidelines for adolescent
of the program. At the time of this report—14 months preventive services. Arch Pediatr Adolesc Med 1997;151:958 –9.
4. Klein JD, Graff CA, Santelli JS, Hedberg VA, Allan MJ, Elster AB. Developing
later—four of the five practices continue to use the quality measures for adolescent care: validity of adolescents’ self-reported
adolescent PDA screeners. receipt of preventive services. Health Serv Res 1999;34(1Pt2):391– 404.
5. Ma J, Wang Y, Stafford RS. U.S. adolescents receive suboptimal preventive
counseling during ambulatory care. J Adolesc Health 2005;36:441.
Conclusion 6. Klein JD, Wilson KM, McNulty M, Kapphahn C, Collins KS. Access to medical
care for adolescents: results from the 1997 Commonwealth Fund Survey of the
This study suggests that clinicians can be effective Health of Adolescent Girls. J Adolesc Health 1999;25:120 –30.
7. Knight J, Harris S, Sherritt L, et al. Adolescents’ preferences for substance
change agents when provided efficient screening tools abuse screening in primary care practice. Subst Abus 2007;28:107–17.
and training. After a well visit, reported exercise levels 8. American Academy of Pediatrics. Periodic Survey of Fellows. 1998.
improved in the intervention group and declined with- www.aap.org/research/periodicsurvey/ps31a.htm.
9. Glasgow RE, Vogt TM, Boles SM. Evaluating the public health impact of
out the Healthy Teens approach. The importance of health promotion interventions: the RE-AIM framework. Am J Public
assessing motivation and customizing counseling is Health 1999;89:1322–7.
supported. Further large-scale research is needed that 10. Glasgow RE, Magid DJ, Beck A, Ritzwoller D, Estabrooks PA. Practical
clinical trials for translating research to practice: design and measurement
links the effectiveness of the clinician in motivating the recommendations. Med Care 2005;43:551–7.
adolescent with later patient outcomes. 11. Olson AL, Gaffney CA, Hedberg VA, et al. The healthy teen project: tools
to enhance adolescent health counseling. Ann Fam Med 2005;3(S):S63–5.
12. Resnicow K, Davis R, Rollnick S. Motivational interviewing for pediatric
This study was funded by grant #53765 from the Robert Wood obesity: conceptual issues and evidence review. J Am Diet Assoc 2006;
Johnson Foundation. 106:2024 –33.
The authors would like to thank the staff and clinicians in 13. Erickson SJ, Gerstle M, Feldstein SW. Brief interventions and motivational
the participating Clinicians Enhancing Child Health prac- interviewing with children, adolescents, and their parents in pediatric health
care settings: a review. Arch Pediatr Adolesc Med 2005;159:1173– 80.
tices: Upper Valley Pediatrics (Bradford VT); Exeter Pediatric 14. Rollnick S, Mason P, Butler C. Health behavior change: a guide for
Associates (Exeter NH); Doctors Who Care (Enfield NH); practitioners. New York: Churchill Livingstone, 1999.
New London Pediatric Care Center (Georges Mills NH); and 15. Nigg CR, Courneya KS. Transtheoretical model; examining adolescent
the Robert Mesropian Community Health Center (Lebanon exercise behavior. J Adolesc Health 1998;22:214 –24.
16. Patrick K, Sallis JF, Prochaska JJ, et al. A multicomponent program for
NH). Special thanks to our programmer, Dr. Zsolt Nagalydi,
nutrition and physical activity change in primary care: PACE⫹ for adoles-
for his role in developing the PDA screener. cents. Arch Pediatr Adolesc Med 2001;155:940 – 6.
No financial disclosures were reported by the authors of 17. Hollis JF, Polen MR, Whitlock EP, et al. Teen reach: outcomes from a
this paper. randomized, controlled trial of a tobacco reduction program for teens seen
in primary medical care. Pediatrics 2005;115:981–9.
18. Gortmaker SL, Peterson K, Wiecha J, et al. Reducing obesity via a
school-based interdisciplinary intervention among youth: Planet Health.
Arch Pediatr Adolesc Med 1999;153:409 –18.
References 19. Gadomski A, Bennett S, Young M, Wissow LS. Guidelines for adolescent
1. Klein JD, Matos Auerbach M. Improving adolescent health outcomes. preventive services: the GAPS in practice. Arch Pediatr Adolesc Med
Minerva Pediatr 2002;54:25–39. 2003;157:426 –32.

S364 American Journal of Preventive Medicine, Volume 35, Number 5S www.ajpm-online.net