Training Health Care Professionals to Manage Overweight Adolescents: Experience in Rural Georgia Communities


  • For further information, contact: David A. Dennison, MS, CHES, Centers for Disease Control and Prevention, 4770 Buford Hwy, NE, MS K-24, Atlanta, GA 30341; e-mail


ABSTRACT: Context:The obesity epidemic threatens the present and future health of adolescents in the United States. Yet, health care providers lack specific training for pediatric obesity assessment and management. Purpose: This study examined the adherence of rural Georgia primary care practitioners to an overweight adolescent management protocol. The study also documented the prevalence of obesity-associated physiological and behavioral risk factors among overweight adolescent patients. Methods: Ten rural clinics (58 providers) were recruited and received a 90-minute adolescent overweight assessment and management training session. Select biochemical, dietary, physical activity, and physical inactivity behaviors were assessed in overweight adolescent patients. Medical charts were abstracted to assess practitioner compliance with an overweight assessment protocol and patient adherence to a 16-week follow-up visit. Findings: Providers were receptive to training and complied with the recommended protocol. Eighty-five overweight adolescents were assessed, but only 49 (57%) completed the scheduled 16-week follow-up visit. Physical, biochemical, and behavioral assessments revealed that 13%-27% of the participants had abnormal levels of lipids, fasting glucose, and glucose/insulin ratio, and 80.5% had waist circumferences above the 90th percentile. Conclusions: Practitioners complied with the assessment and follow-up protocol, leading to the discovery of previously unrecognized risk factors in many overweight adolescent patients. Lack of patient adherence to follow-up was the greatest limiting factor for obesity management. Further efforts are needed to implement and evaluate training to improve the management of adolescent overweight, especially in rural communities.

Overweight among children and teenagers more than tripled between the 1960s and 2002.1 Childhood overweight is not only increasing, but is also becoming more severe.2 The obesity epidemic is linked to the emergence of type 2 diabetes among adolescents,3 and recently has been estimated to account for 6% of newly diagnosed diabetes in non-Hispanic white youth and as much as 76% of newly diagnosed diabetes in American Indian youth.4 This increase in adolescent overweight and diabetes is largely attributable to lifestyle factors,5,6 specifically, physical activity and nutrition.7–9 The link between childhood overweight and other chronic diseases is also increasingly recognized.5,10,11 Obesity in children increases their risk for becoming obese adults, and for developing diabetes, hypertension, coronary heart disease, osteoarthritis, and some cancers.5,10–13

In Georgia, 33% of middle school students and 26% of high school students were found to be at risk of overweight (85th-94th body mass index [BMI] percentile) or overweight (≥95th BMI percentile)14 based on self-reported height and weight and using criteria recommended by the Centers for Disease Control and Prevention (CDC).15 Moreover, the prevalence of overweight in rural settings may be higher than in urban areas.16 Overweight prevalence based on measured height and weight in a statewide sample of Georgia elementary, middle, and high school students was higher in rural (39.9%) and suburban (39.5%) communities than in urban areas (32.5%).17

Most successful approaches to reduce childhood overweight delivered through medical providers have been treatment-oriented and long-term in nature, and have provided a combination of dietary, physical activity, and behavioral therapy.18 However, this type of intensive, long-term counseling is expensive and difficult to provide, especially in rural settings where access to health care is often more limited.16 It is important, therefore, to develop and evaluate training and materials for the management of childhood obesity that are feasible for application in more routine clinical settings, especially in rural communities.

A pediatric obesity training protocol, previously developed, delivered, and tested in a managed care setting, was found to be accepted by practitioners and effective in changing practice patterns and addressing clinician barriers.19,20 The purpose of this study was to examine adherence to this training protocol by rural Georgia primary care practitioners and to assess the compliance of overweight adolescent patients. In addition, this study reports the prevalence of obesity-associated risk factors in overweight adolescents seen in rural Georgia clinics.


Study Design Ten pediatric clinics in rural Georgia were recruited through the Georgia chapter of the American Academy of Pediatrics and agreed to receive the training in obesity assessment. Rural status was defined by the Office of Rural Health Policy's geographic eligibility for rural health programs. Provider adherence to training recommendations and prevalence of obesity-associated risk factors in overweight adolescent patients were assessed via medical chart abstraction 16 weeks post-training. The study protocol was approved by the Institutional Review Board at the ILSI Research Foundation Center for Health Promotion.

Overweight Management Training The training in obesity assessment summarized recommended assessment practices21,22 and included interactive participation in case studies as utilized by the CDC in their growth chart training modules.23 Trainings lasted approximately 90 minutes and were delivered at the clinic site after work hours, with dinner or snacks provided. All the respective office staff were invited to attend the trainings, but in most cases the nurses, nurse practitioners, and physicians attended, while medical assistants and clerks did not. Two individuals delivered the trainings: a pediatrician with obesity management and research experience, and a certified health education specialist.

The specific recommendations included: (1) assess BMI in all patients, (2) prescribe a blood panel assessment for patients with a BMI ≥95th percentile or with a BMI >85th percentile plus a family history of cardiovascular disease or diabetes, and (3) provide behavioral management counseling to patients with behavioral risk factors.

Participants and Measurements Providers at each of the 10 clinics were asked to recruit 10 overweight adolescents from their existing client pool and to assess nutrition and physical activity-related behaviors. These behaviors were assessed utilizing a self-report instrument developed previously, the Nutrition and Activity Self-History Form.19,20 Selected patients were 13-18 years of age and had a BMI ≥95th percentile or a BMI >85th percentile plus a family history of either cardiovascular disease or diabetes. Medical charts from these individuals were abstracted.

Outcome Variables BMI assessment and blood panel prescriptions were used as proxies for adherence to the training protocol. Physical measurements, including weight, height, waist circumference, and blood pressure were performed by clinic nurses using office scales and height measuring equipment. The prevalence of waist circumference above the 90th percentile24 and blood pressure above the 95th percentile25 was used to define health risk related to these measures. Blood chemistries included: total cholesterol, HDL, LDL, triglycerides, glucose, insulin, and hemoglobin A1c. These measurements were assessed at baseline and a follow-up visit at week 16.

Statistical Analysis Descriptive statistics were used to assess provider participation, adherence to the training protocol, physical and blood chemistry measurements, and the nutrition and physical activity behaviors of the adolescents. There were no significant differences in findings based on race, gender, or age and, therefore, these data are not reported. Statistical test were done using SPSS for Windows, version 14.0 (SPSS, Inc., Chicago, Ill).


Implementation, Adherence, and Compliance to Training All medical providers in the rural clinics, a total of 58, participated in the training. During the interactive sessions, 56 of the 58 providers (97%) reported that they had not previously utilized BMI or BMI-for-age percentile to assess childhood overweight. The 10 participating clinics enrolled a total of 86 overweight adolescents meeting study criteria. Of those enrolled, 49 (57%) were African American, 32 (37%) were white, and 5 (6%) were Hispanic. Forty-eight (56%) subjects were female and 38 (44%) were male. Subjects ranged in age from 13 to 18 years, with a mean age of 14.5 years.

Based on chart review, 85 of the 86 (99%) adolescents had a recorded BMI assessment. None of the adolescents had a charted BMI percentile prior to the date of training. Additionally, all 86 were prescribed a fasting blood panel and 74 of the 86 (86%) adolescents completed the fasting blood chemistry evaluation. Medical charts revealed that none of these adolescents had previously been prescribed a blood panel for overweight or obesity evaluation. All 86 participating adolescents were asked to return 16 weeks later for a follow-up visit, but only 49 (57%) actually completed the follow-up visit.

Prevalence of Obesity-Related Risk FactorsTable 1 displays baseline physical and biochemical measurements and risks assessed in these overweight adolescents (at risk determinations were based on analysis with HealthWatch Pro 2.0, BV Tech Inc., Martinez, Ga). Results indicate a high prevalence of obesity-related risk factors for metabolic diseases, such as diabetes and cardiovascular diseases, in the participants. In particular, 80.5% had waist circumference over the 90th percentile, indicating a high level of central adiposity; 33.7% had systolic blood pressure over the 95th percentile, and 15.6% had diastolic blood pressure over the 95th percentile adjusted for age and height. Abnormal levels of lipids, fasting glucose, and glucose/insulin ratio were observed in 13%-27% of the sample.

Table 1.  Baseline Physical, Biochemical, and Risk Assessment Measures
 NMeanSEAt Risk
Physical measurements
 BMI (kg/m2)8536.460.8>95th percentile7993
 BMI Z-score852.250>2.06576
 Waist circumference (in.)8243.140.8>90th percentile6681
Blood pressure
 Systolic83119.21.5>95th percentile2834
 Diastolic8374.851>95th percentile1316
Blood chemistries
 Triglycerides (mg/dL)7393.079.6>1501014
 Total cholesterol (mg/dL)74160.83.8>2001216
 HDL cholesterol (mg/dL)7444.341.4 <351419
 LDL cholesterol (mg/dL)491054.7>1301020
 Fasting glucose (mg/dL)7092.261.2>1001319
 Fasting insulin (pmol/L)6421.572 >6023.1
 Glucose/insulin ratio606.841  <71127
 Hgb A1c (mmol/L)615.170.1  >700

Nutrition and Physical Activity Behaviors A total of 61 (71%) adolescents responded to the Nutrition and Activity Self-History questionnaire. Table 2 summarizes the reported dietary behaviors. Thirty-eight percent reported consuming vegetables 0-1 time per day. Almost 60% reported consuming fruits less than 2 times per day. Additionally, 4 or more times per day, 12% consumed fried foods, 25% consumed sweet or salty snacks, and 45% reported consuming soda and/or sweetened beverages.

Table 2.  Percent of Adolescents Reporting Selected Dietary Behaviors (n = 61)
 Times per DayTotal
Consumes vegetables385642100%
Consumes fruits583840100%
Consumes fried foods4048102100%
Consumes sweet or salty snacks2253169100%
Drinks soda/sweetened beverages17382817 100%

The physical activity and sedentary reported behaviors are summarized in Table 3. Over 50% reported participating in physical education only 0-1 days per week while 45% reported accumulating 60 minutes of daily physical activity 4-7 days per week. More than half (59%) reported watching television 3 or more hours per day and 33% reported playing computer or video games 3 or more hours per day. Additionally, 21% reported a combined time of television watching and playing computer or video games greater than 6 hours per day.

Table 3.  Percent of Adolescents Reporting Selected Physical Activity and Sedentary Behaviors (n = 61)
 Days per WeekTotal
Participate in physical education521830n/a100%
Accumulate 60 min. of activity22333411100%
 Hours per DayTotal
<11-23-45 or more
Watch TV10313128100%
Play computer/video games442326 7100%


The study demonstrated a high level of interest and concern among practitioners regarding adolescent obesity, and receptiveness to training to improve their clinical management skills. All 58 providers in the selected clinics attended the trainings, which were held outside of normal work hours, and devoted substantial amounts of time to counseling and patient follow-up visits. The training and participation in the study also appeared to have an impact on practice patterns. Before training, 2 providers reported that they calculated BMI for their patients, and none of the charts had a documented BMI. After training, all but one chart had a recorded BMI, and all charts indicated that the appropriate blood panel had been ordered. In follow-up discussions, providers indicated their intent to continue these practices.

The recommended adolescent obesity assessment and management protocol appears to have been widely adopted and implemented by the clinicians. However, compliance of adolescent patients with the recommended testing and follow-up visits was less complete. Eighty-six percent of the adolescents completed the prescribed fasting blood chemistry evaluation but only 57% returned for their 16-week follow-up visit. The low rate of the follow-up visit was primarily the result of missed appointments, which may have been due to lack of transportation or the inability of parents or caregivers to accompany patients. This was despite repeated attempts of clinic staff to contact and encourage patients to attend and a modest ($10 per visit) financial incentive. Therefore, an important barrier to adolescent obesity management in this rural setting appears to be limited compliance by adolescent patients with follow-up visits.

The implementation of the recommended assessment procedures led to the discovery of numerous obesity-related risk factors. Baseline physical measurements indicated a high rate of abdominal obesity as reflected in the high proportion of subjects with waist circumference measurements above the 90th percentile. These findings are significant since abdominal obesity may be a better predictor than overall obesity of the risk for cardiovascular disease and type 2 diabetes.26 Elevated blood pressures, blood lipids, and fasting glucose values also demonstrated a significant level of health risk among these overweight adolescents.

The self-reported dietary, physical activity, and sedentary behaviors indicated a low consumption of fruits and vegetables and a high consumption of high caloric products, similar to national data.27 These overweight adolescents also reported less than the recommended level of physical education participation. Consistent with a national sample, only 30% participate in daily physical education.27 Other studies have also shown overweight adolescents to be less active than normal weight adolescents.28–30 Perhaps most alarming is the percentage of reported sedentary time. Sedentary behavior is much higher in these rural, overweight adolescents compared to a national sample. Thirty-three percent of participating overweight adolescents reported playing computer or video games for 3 or more hours per day, compared with 21.1% in a national sample.27 Likewise, 59% in this population reported watching television 3 or more hours per day versus 37.2% in the national sample.27 These behavior patterns place these adolescents at risk to remain overweight or become even more overweight in the future.


This pilot study was designed to address barriers to and management of pediatric overweight by exploring adherence of rural Georgia clinicians to a protocol.19–22 Results demonstrate that clinic staff were highly responsive to training and readily implemented routine BMI screening and biochemical evaluation according to recommended guidelines. Overweight adolescents' lack of compliance with follow-up visits appears to be a significant limiting factor for obesity management for these rural communities. Physical and biochemical measures in these overweight adolescents indicate a high prevalence of obesity-related risk factors. These findings indicate the need for further efforts to implement and evaluate training to improve the management of adolescent overweight, including understanding adolescent readiness and barriers to compliance, especially in rural communities where the prevalence of obesity may be higher and health care less accessible.