Abstract
Objective. To examine clinician-reported diagnosis of obesity and frequency of blood pressure assessment and diet and exercise counseling during ambulatory visits made by children and adolescents.
Methods. The National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey 1997 to 2000 were combined for visits to clinicians of 2- to 18-year-olds. Well-child visits (WCVs) were examined for frequencies of obesity diagnosis, blood pressure screening, and diet and exercise counseling in relation to patient and clinician characteristics. Multivariate models examined the relationship of patient and visit characteristics with diet and exercise counseling.
Results. Of the 32 930 ambulatory visits made by 2- to 18-year-olds in 1997–2000, obesity was diagnosed at 0.78% of all visits and 0.93% of WCVs. Blood pressure assessment was reported in 61.1% of WCVs with obesity diagnosis compared with 43.9% of WCVs without obesity diagnosis. WCVs with obesity diagnosis had higher diet counseling rates (88.4% vs 35.7%) and higher exercise counseling rates (69.2% vs 18.6%). Diet counseling was reported for 88.4% and exercise counseling was reported for 69.2% of visits with an obesity diagnosis compared with 35.7% and 18.6% during WCVs without a diagnosis of obesity. In multivariate analyses, factors associated with diet counseling at WCVs were diagnosis of obesity (odds ratio [OR]: 12.9; 95% confidence interval [CI]: 3.0–55.3), being seen by pediatricians (OR: 2.5; 95% CI: 1.6–3.9), 2- to 5-year-olds compared with 12- to 18-year-olds (OR: 0.7; 95% CI: 0.5–1.0), and self-pay compared with private insurance visits (OR: 0.6; 95% CI: 0.4–0.9). Associations with exercise counseling were similar to those for diet counseling, but exercise counseling occurred less frequently in visits by black youths compared with white youths (OR: 0.5; 95% CI: 0.3–0.8).
Conclusions. Clinicians may overlook obesity during WCVs. Programs to increase obesity diagnosis could improve diet and exercise counseling rates, but even with diagnosis of obesity, significant opportunities for screening and intervention are missed.
In 2001, an estimated 111 million visits were made to office-based clinicians by children and adolescents.1 At the same time, 15% of youths from 6 to 19 years of age were overweight/obese, on the basis of the recommended cut point of the 95th percentile BMI.2 These estimates indicate that obesity is rapidly becoming the most common chronic medical condition affecting children and adolescents.3,4 The high prevalence of childhood obesity is associated with increasing rates in childhood of conditions that almost exclusively have been seen in adults until recently, such as type 2 diabetes5 and the metabolic syndrome.6 As a result, America's pediatric clinicians must address with increasing frequency obesity and its associated comorbidities.
In 1998, the Maternal and Child Health Bureau, Health Resources and Services Administration, Department of Health and Human Services, convened an expert panel to recommend guidelines for the evaluation and treatment of obesity in children and adolescents.7 These recommendations included how to identify childhood obesity, how to assess medical risks associated with obesity and readiness for healthy lifestyle modifications, and how to guide child and family behavior change to improve eating and activity habits. The American Academy of Family Physicians has supported a similar guideline.8
To assess the impact of these recommendations on clinical practice and to identify additional educational needs, a needs assessment survey was distributed to >1000 members of the American Academy of Pediatrics, 800 members of the National Association of Pediatric Nurse Associates and Practitioners, and 1600 members of the American Dietetic Association.9 Seventy-three percent to 88% of pediatric clinicians reported that they “often” or “always” made recommendations for weight control when they saw overweight children or adolescents. Many clinicians, however, reported lack of patient motivation, lack of time available for counseling, lack of effective treatment, and lack of reimbursement as frequent barriers to the treatment of childhood obesity.10 Although the majority of respondents indicated that they routinely address excess weight in overweight youths, with <20% of pediatricians surveyed completing the study, the sample may not have been representative.
We sought another method to examine the frequency with which clinicians document obesity, medical evaluation, and lifestyle counseling that is recommended as components of appropriate care of obese youths. The aims were to use a nationally representative sample of outpatient visits by children and youths to determine how often clinicians note obesity as a diagnosis, assess blood pressure, and provide diet and exercise counseling and to examine factors associated with these aspects of obesity-related health care. The focus of these analyses was well-child visits (WCVs), because anticipatory guidance and health screening for chronic conditions are recommended during these encounters.11,12
METHODS
Data Source
The National Center for Health Statistics measures various aspects of ambulatory health care utilization with the National Ambulatory Medical Care Survey (NAMCS) and the National Hospital Ambulatory Medical Care Survey (NHAMCS). NAMCS, an ongoing survey since 1972, collects information about patient visits to private, non–hospital-based offices, clinics, and health maintenance organizations of non–federally employed US physicians. The NHAMCS, a similar survey begun in 1992, includes visits made to outpatient departments of nonfederal, short-stay hospitals (hospitals with an average stay of <30 days). These public access data sets, when weighting estimates are applied, are nationally representative of ambulatory visits across the United States.
Participating clinicians complete a 1-page patient record form for a systematic, random sample of ambulatory care visits during a randomly assigned week. Clinicians or the office staff record demographic information about the child, reasons for visit, diagnoses, prescriptions written, and selected screening and diagnostic procedures. The survey instructs the clinician to list a primary diagnosis and up to 2 additional diagnoses per visit and to include any chronic conditions, eg, depression, obesity, asthma. The simple fill-in-the-blank responses for diagnoses are later translated into International Classification of Diseases, Ninth Revision (ICD-9) codes by the National Center for Health Statistics staff. Information on patient treatment, including counseling or education provided, requires only check-box responses. Each year, the NAMCS and Outpatient NHAMCS encounter forms are identical. Minor changes are made to both survey forms biannually, but the current analyses include only questions that did not change during 1997–2000. Data from 2001–2002 were not used because of significant wording changes in key items.
Data from the 1997 to 2000 NAMCS and NHAMCS were combined, and emergency department visits were excluded, leaving ∼90% of child and adolescent visit data from NAMCS and the remainder from the hospital outpatient departments reported in NHAMCS.13 Inclusion of NHAMCS data increased the number of pediatric visits available for final analysis and also avoided excluding children and youths who use hospital-based outpatient departments.
Independent and Dependent Variables
Visits of patients who were 2 to 18 years of age and who were seen by a physician or by a midlevel provider such as a nurse practitioner or physician's assistant were included. Visits were classified as having a diagnosis of obesity when ICD-9 codes for obesity (278.00), morbid obesity (278.01), or excess weight gain (783.1) were used. WCVs and annual visits were identified in the clinician diagnosis section on the basis of ICD-9 codes (V20-, V70-). Covariates included patient demographic characteristics (age, gender, race, and ethnicity), insurance type, type of clinician seen (physician or midlevel practitioner), and physician specialty (pediatric, general or family practice, and other). Insurance type was categorized as private insurance, Medicaid, self-pay, or other. The dependent measures included rates of blood pressure screening and rates of diet and exercise counseling at ambulatory visits.
Analysis
χ2 tests were used to test for differences in proportions. Associations with diet or exercise counseling that were significant (P ≤ .10) in bivariate analyses were placed in a logistic regression model to determine independent associations with diet and exercise counseling. All percentages in this study were weighted to reflect national estimates. SUDAAN software (Research Triangle Institute, Research Triangle Park, NC) was used to account for the complex sample design.
RESULTS
There were 32 930 visits of 2- to 18-year-olds from 1997 to 2000, representing 131.2 million visits over the 4 years when weighted national estimates are applied. Among these visits, 281 (0.78%) carried an ICD-9 code diagnosis of obesity, morbid obesity, or excess weight gain. Among WCVs, the frequency of obesity diagnosis was 0.9%. An obesity diagnosis was included more often at visits to the middle-age group (6- to 11-year-old children) and at visits to pediatricians (Table 1).
Child and Clinician Characteristics Associated With a Diagnosis of Obesity at WCVs (N = 3514 Visits): NAMC and NHAMC Surveys, 1997–2000
Table 2 shows the prevalence of blood pressure, diet, and exercise counseling at WCVs. The rate of blood pressure screening was 43.9% among WCVs without obesity identified and was 61.1% when obesity was identified, but the difference was not statistically significant. Blood pressure screening increased with age group (P < .001) and also differed by race and insurance status: 47.7% of visits of white children as compared with 29% of visits of black children (P = .01), 62.2% of other insurance, 46.4% of self-pay, 44.9% of private insurance, and 34.3% of Medicaid (P = .02). Blood pressure screening occurred in 62.8% of visits with general/family practitioners compared with 36.6% of pediatricians (P < .001).
Rates of Blood Pressure Screening, Diet Counseling, and Exercise Counseling at WCVs (N = 3514): NAMC and NHAMC Surveys, 1997–2000
Diet counseling occurred in 88.4% of WCVs with an obesity diagnosis, compared with 35.7% of visits with no obesity diagnosis (P = .002; Table 2). Exercise counseling occurred more frequently when obesity was diagnosed (69.2%) compared with visits without obesity diagnosis (18.6%; P = .02). The rate at which clinicians reported diet counseling differed significantly by age group (P < .002) and insurance type (P = .01), and pediatricians reported providing diet counseling more frequently (43.5%) than did general practice physicians (21.4%) or other physicians (9.3%; P < .001). Exercise counseling rates also varied by age, insurance type, and clinician type. Exercise counseling was reported in 20.4% of visits with pediatricians, 15.9% of visits with general/family practitioners, and 5.6% of other physicians (P = .01). Although diet-counseling rates did not differ by race, exercise counseling occurred less frequently among black children (P = .03).
In a multivariate model, a diagnosis of obesity was independently associated with diet counseling (odds ratio [OR]: 12.9; 95% confidence interval [CI]: 3.0–55.3) and exercise counseling (OR: 10.5; 95% CI: 2.7–40.7; Table 3). The odds of diet counseling were 2.5 times greater among visits to pediatricians than visits to family practitioners. Visits for adolescents were 30% less likely to include diet counseling compared with visits for 2- to 5-year-olds, and diet counseling occurred 40% less often when the source of payment was self-pay compared with private insurance. Similar associations were found for exercise counseling, except that 6- to 11-year-olds and 12- to 18-year-olds had higher odds of exercise counseling. In addition, black children were less likely than white children to receive exercise counseling (OR: 0.5; 95% CI: 0.3–0.8).
Factors That Are Independently Associated With Diet and Exercise Counseling at WCVs Among Children and Youths 2–18 Years of Age (N = 3514): NAMC and NHAMC Surveys, 1997–2000
DISCUSSION
This is the first study, to our knowledge, to use a nationally representative sample of outpatient visits to investigate rates of diagnosis of obesity in children and adolescents, screening for hypertension and nutrition and exercise counseling. For each of these domains, the rates were much lower than anticipated. In addition, racial and health insurance–related disparities were apparent.
At the time this survey was conducted, an estimated 15% of children and adolescents in the United States were obese,2 yet obesity was reported as diagnosed in only 0.7% of all visits and 1% of WCVs of children aged 2 to 18 years. Although clinicians may have recognized that the child was obese and discussed this condition with the family in some visits without listing obesity as a diagnosis, the marked discrepancy between diagnosis and national prevalence suggests that this condition was overlooked. A recent report found that pediatricians are more concerned by obesity on the basis of BMI charts compared with height and weight growth charts.14 Physicians have also been shown to document obesity in children at lower levels of obesity.15 Screening for blood pressure may have been higher when obesity was diagnosed, but the low rates of obesity diagnosis limited power to detect difference. However, blood pressure rates in all WCVs were still well below the 90% or more reported in other surveys16,17 and below the recommendations to screen at all WCVs beginning at 3 years of age.11,12 Although not shown here, cholesterol screening rates were higher when obesity was diagnosed. Because the appropriateness of cholesterol screening depends on unavailable information, including family history and previous screening, we did not examine associations of cholesterol screening with other variables.
In the mid-1980s, pediatricians reported a lack of comfort counseling or following up children and adolescents on lifestyle changes for cardiovascular disease prevention.18,19 The low diet and exercise counseling rates at WCVs may reflect physicians' reluctance to address conditions for which they believe they have little impact. The low counseling rates among black children are especially disturbing because the higher obesity prevalence in this group should lead to higher rates.2–4 Instead, rates of dietary counseling were similar for visits by black and white children, and exercise counseling occurred half as often in visits by black children.
This study has several limitations. In the absence of measured height and weight data, we do not know the validity of the obesity diagnosis or the characteristics of children who have excess weight and whose visit lacked an obesity diagnosis. Physicians may vary in their use of the term “obesity,” reserving it for more severe excess weight, in line with the adult definitions, rather than identifying all children with BMI ≥95th percentile as “overweight” or “obese,” as recommended by the Centers for Disease Control and Prevention and others. The translation of diagnoses, provided by physicians in standard medical terminology, into ICD codes could add variation to the definition of obesity. However, the ICD-9 code book definition for obesity is straightforward. The unit of analysis is the visit and not the patient, and it therefore is possible that obese children first were identified at a previous visit. Although documentation may have been lacking at the time of the survey, the chronic nature of obesity suggests that it should be discussed annually. Although clinicians might have more time to address the condition at a separate visit,11 lack of reimbursement20 is an important barrier to a focused obesity visit.9,17,21 Just as clinicians do not document all practice behavior in medical charts,22 the participating clinicians might have addressed obesity but not included it as a diagnosis through oversight, habit, or lack of space on the form if >3 conditions were addressed during the visit. Despite the inclusion of 4 years of surveys, the low number of WCVs with obesity diagnoses limited the power to test for additional associations with demographic factors and screening and counseling practices. Counseling may be underreported, as suggested by a study that used the NAMCS form to assess its validity against visit observation.23 Finally, these survey instruments demonstrate simple frequencies of reported counseling at the time of the visit and do not capture the quality of counseling provided.
The NAMCS/NHAMCS data sets have an established record of providing useful information about clinical practice for a variety of pediatric health problems,24–30 and this analysis describes a universe of >130 million outpatient encounters, yet <1% of visits included a diagnosis of this major epidemic condition. When obesity was in fact diagnosed, the rates of screening for comorbid conditions and the rates of dietary and exercise counseling increased, which suggest that programs to promote diagnosis would result in improved lifestyle counseling rates. Moreover, if the racial and health insurance–related disparities in both obesity diagnosis and management are confirmed, then targeted programs to reduce the disparities will be important pieces of the health care system's response to this problem.
Acknowledgments
Dr Cook is supported by Faculty Development in Primary Care, Health Resources and Services Administration grant T32PE12002, and Dr Barlow is supported by Agency for Healthcare Research and Quality grant K08 HS013901-01.
Footnotes
- Accepted October 14, 2004.
- Reprint requests to (S.C.) University of Rochester School of Medicine and Dentistry, Department of Pediatrics, 601 Elmwood Ave, Box 278881, Rochester, NY 14642. E-mail: stephen_cook{at}urmc.rochester.edu
This paper was presented in part at the North American Association for the Study of Obesity; October 12–14, 2003; Ft Lauderdale, FL; and at the annual meeting of the Pediatric Academic Society; May 3–6, 2003; Seattle, WA. No conflict of interest declared.
REFERENCES
- Copyright © 2005 by the American Academy of Pediatrics
In this issue
Jump to section
Related Articles
Cited By...
- Are Pediatricians Diagnosing Obese Children?
- FitwitsTM Leads to Improved Parental Recognition of Childhood Obesity and Plans to Encourage Change
- American Heart Association Childhood Obesity Research Summit Report
- Preventive Care for Adolescents: Few Get Visits and Fewer Get Services
- Use of an Electronic Medical Record System to Support Primary Care Recommendations to Prevent, Identify, and Manage Childhood Obesity
- Acanthosis Nigricans and Diabetes Risk Factors: Prevalence in Young Persons Seen in Southwestern US Primary Care Practices
- Identification of Overweight Status Is Associated With Higher Rates of Screening for Comorbidities of Overweight in Pediatric Primary Care Practice
- Screening and Counseling for Obesity in the Ambulatory Care Setting: In Reply
- Screening and Counseling for Obesity in the Ambulatory Care Setting