- Email: [email protected]
Obesity and eating disorders in older adolescents: Does early dieting make things better or worse?
152
Abstracts / 36 (2005) 111-154
for example: increased fruit & vegetable consumption (1.6 to 2.6 servings/day, p ⫽ .03), decreased TV watching (2.8 to 2.3 hrs/day, p ⫽ .05), fewer pts eating fast food items ⱖ 3 days/wk (30% to 5%, p ⫽ .04). A higher percentage of involved pts lost wt (43%) than uninvolved patients (24%, p ⫽ .05) although, on average, pts from both groups gained wt (⫹0.24 lbs/mo vs. ⫹0.59 lbs/mo, NS) over a follow-up period that averaged 9 mos. Involved pts were significantly heavier than uninvolved pts (mean BMI 40.1 vs. 34.4 kg/m2). Pts who lost wt had significantly lower fasting insulin levels than those who did not. Conclusions: An intervention carried out in 3 SBHCs showed promise in modifying risk factors for type 2 DM among Hispanic and African-American high school students at high risk due to obesity. Support: Chicago Community Trust, Washington Square Health Foundation.
PII: S1054-139X(04)00384-2 81.
EMOTIONAL EATING IN ADOLESCENT FEMALES Cathleen Steinegger, M.D., Lorah Dorn, Ph.D., Cynthia Goody, Ph.D., R.D., L.D.#, Philip R. Khoury, M.S., and Stephen R. Daniels, M.D., Ph.D. Cincinnati Children’s Hospital Medical Center and the #University of Cincinnati Medical Center, Cincinnati, Ohio. Purpose: To investigate the emotional eating (EE) behavior of white and black adolescent females across puberty and adolescence. The overall aims include an examination of the patterns and predictors of EE, the relationship between the degree of EE and the average daily caloric intake, and whether EE influences final BMI. This report focuses on age and race-related patterns of EE in adolescents. Methods: This is a secondary data analysis of the National Heart, Lung and Blood Institute Growth and Health Study (NGHS) which collected data annually from 1987 through 1997 on black and white girls, beginning at 9 to 10 years of age. A subset of 870 girls, 431 white and 439 black, from one site was used in this analysis. Longitudinal data included: height, weight, pubertal stage (by exam), demographic data, 3-day food diaries, and the EmotionInduced Eating Scale (EIES) developed for the NGHS. The EIES is a 7 item scale with possible scores ranging from 7 (low EE) to 21 (high EE). Results: Using SAS®, a GEE model was used for analysis. The EE levels were low and, overall, age did not influence the EE score. However, at 9 –10 years, black girls had significantly higher levels of EE than white girls (10.4 vs. 9.8, p ⫽ 0.002). There was a significant age-by-race interaction with white girls increasing their EE score by 0.16 for each 1 year increase in age (p ⬍0.0001), while black girls decreased their EE score by 0.059 (p ⫽ 0.006) with each year. By study termination, 18 –19 years of age, the black girls’ were reporting lower EE scores than the white girls (9.5 vs. 10.2, p ⬍.001). Conclusions: The pattern of EE through adolescence varies by race. While black girls begin the adolescent years with higher EE levels than white girls, by 19 years of age, the races are reversed and white girls identify more EE behaviors than blacks. By identifying different patterns and influences on eating behavior between races, we may be able to create more effective nutrition interventions for disease prevention and treatment. Our subsequent analyses will explore the influence of puberty and pubertal timing,
depression, and socioeconomic status on the EE behavior followed by the role of EE on caloric intake and body mass.
PII: 1054-139X(04)00385-4 82.
OBESITY AND EATING DISORDERS IN OLDER ADOLESCENTS: DOES EARLY DIETING MAKE THINGS BETTER OR WORSE? Dianne Neumark-Sztainer, Ph.D., M.P.H., R.D. Melanie M. Wall, Ph.D. Marla E. Eisenberg, Sc.D., M.P.H. Mary Story Ph.D., R.D. University of Minnesota School of Public Health and School of Medicine. Minneapolis, Minnesota. Purpose: Eating and weight-related problems are of major public health concern owing to their high prevalence in adolescents and the potentially serious physical and psychosocial consequences. While dieting is commonly used as a weight loss strategy, we hypothesized that it may be exacerbating the problems of both obesity and eating disorders. The purpose of this study is to examine the prevalence of a range of weight-related problems in a large sample of older adolescents and to explore longitudinal associations between dieting in high school and weight gain, use of disordered eating behaviors, and the onset of eating disorders five years later in young adulthood. Methods: Project EAT was originally conducted in 1998 –1999 on an ethnically and socioeconomically diverse sample of middle school and high school students. In 2003–2004, participants were contacted and mailed surveys to assess eating and weight-related issues. This analysis focuses on the 1729 older adolescents who were followed from high school to young adulthood (Mean age ⫽ 20.4, SD ⫽ 0.89). Descriptive proportions are stratified by gender. Logistic regression is used to estimate odds ratios and 95% confidence intervals relating early dieting to later weight-related problems, adjusting for early weight-related problems. Results: Preliminary analyses show that 24% of the females and males were overweight or at-risk for overweight (BMI⬎ 85th percentile), using self-reported height and weight data. Binge eating was reported by 15.6% of females and 4.6% of males. Extreme weight control behaviors (vomiting, laxatives, diet pills, and diuretics) were reported by 24.3% of females and 6.6% of males. The presence of an eating disorder in the past year was reported by 5.7% of females and 0.5% of males. Preliminary analyses show that early dieting was significantly associated with becoming overweight in young adulthood (females: OR: 1.95, CI: 1.29,2.96; males: OR: 1.88, CI: 1.19,2.97) and with the later onset of binge eating (females: OR: 2.2; CI: 1.4,3.5; males: OR: 2.8; CI: 1.8, 6.0). Early dieting was associated with the onset of extreme weight control behaviors in females (OR: 2.2, CI: 1.6,3.2) and associations in males were of borderline significance (OR: 1.9; 1.0, 3.5). Early dieting was also associated with eating disorders in the past year in females (OR: 3.1, CI: 1.5,6.4) but associations in males were not statistically significant (OR: 2.3, CI: 0.4,14.0). Conclusions: In order to develop interventions that have the potential to prevent the broad spectrum of weight-related problems, we need to identify modifiable shared risk factors. Our findings suggest that dieting places youth at risk for eating disorders, disordered eating behaviors, and obesity. Efforts to prevent dieting, and, instead, encourage the adoption of long-term healthy
Abstracts / 36 (2005) 111-154
153
eating and physical activity behaviors are needed with clinics, schools, and in the broader community. Support: Grant # R40 MC 00319 – 02 from the Maternal and Child Health Bureau, Health Services Research Administration, U.S. Department of Health and Human Services (NeumarkSztainer, PI).
physiologic parameter may stabilize later than temperature and heart rate in adolescents recovering from malnutrition due to an eating disorder. Support: Partially supported by the Health and Resource Services Administration, Maternal and Child Health Bureau, Grant # 6 T71 MC00011.
PII: S1054-139X(04)00386-6
PII: S1054-139X(04)00388-X
83.
84.
INPATIENT TREATMENT OF ADOLESCENT PATIENTS WITH EATING DISORDERS: MEDICAL AND FINANCIAL OUTCOMES Jennifer L. Kalisvaart, M.D., Albert C. Hergenroeder, M.D., Section of Adolescent Medicine and Sports Medicine, Department of Pediatrics, Baylor College of Medicine, Houston, Texas.
VARIATIONS IN ADMISSION PRACTICES FOR ADOLESCENTS WITH ANOREXIA NERVOSA: A NORTH AMERICAN SAMPLE Jonathan Mansbach, MD, Beth Schwartz, BA, Jenna Gordon, MA, Debra K. Katzman, MD, and Sara Forman, MD. Divisions of Adolescent Medicine, Children’s Hospital Boston, Boston, MA, and The Hospital for Sick Children, Toronto, Ontario.
Purpose: To describe the medical outcomes and insurance reimbursement for treatment of patients with eating disorders on an adolescent medical unit. Methods: Using a descriptive, retrospective, cohort study design, we investigated the outcomes at the time of discharge for 39 patients with medical complications of anorexia nervosa (AN) or eating disorder not otherwise specified (EDNOS) following treatment on an adolescent medical unit, admitted over a three-year period. Admission criteria included a combination of estimated ideal body weight (%IBW) ⬍85%, bradycardia (resting heart rate ⱕ 50 bpm), hypothermia (ⱕ 35.5° C), and orthostatic tachycardia (increase ⱖ 35 bpm after standing for 5 minutes), and/or failure of outpatient treatment. 74% (N ⫽ 29) were diagnosed with anorexia nervosa (AN); 26% (N ⫽ 10) with eating disorder not otherwise specified (EDNOS). Outcomes included hospital and professional charges and reimbursements. Paired and independent samples ttests, Pearson Chi-square, and Wilcoxon rank sum analyses were used. Results: The mean age of the patients was 16.1 ⫾ 1.9 years. % IBW at admission was 74.8 %. Average daily weight gain was 0.100 ⫾ 0.06 kg/d. Hospital duration averaged 51 ⫾ 30 days. 64% were bradycardic and 44% were hypothermic at admission that improved to 17% and 0%, respectively, at discharge (p ⬍ .001 for both). Postural orthostatic tachycardia did not improve over the course of treatment. Discharge criteria included achieving 85% IBW. Medical diagnoses, including malnutrition, bradycardia, hypothermia and orthostatic tachycardia, were listed as the admitting and discharge diagnoses. Average total hospital and professional charges were $105,853 per patient, and insurance companies reimbursed an average of 62% of these total charges out of the patients’ medical benefits policies. No patient was discharged based on insurance criteria. Uniformly, the insurance companies denied coverage for the number of requested days. Conclusions: The majority of expenses for inpatient care on this adolescent medical unit were reimbursed by insurance companies under the patients’ medical benefits. However, the reimbursement was inadequate in order to achieve a discharge weight of 85% IBW, a weight associated with improved outcome. Inadequate reimbursement by insurance companies threatens the ability to provide optimal inpatient medical treatment for patients with eating disorders who are medically unstable. In addition, in contrast to what others have reported, normalization of orthostatic pulse may not be a criterion on which to base medical stabilization. This
Purpose: In 2000, the American Psychiatric Association published consensus admission guidelines for patients with eating disorders, but studies are needed to evaluate actual practices for clinical admission and treatment differences among patients with Anorexia Nervosa (AN). The purpose of this study was to assess the variability in admission practices and medical inpatient care for adolescent patients with AN. Methods: Members of the 2001–2004 Eating Disorder Special Interest Group from the Society for Adolescent Medicine were contacted via email and/or telephone. Participants completed a 30-minute structured telephone interview about their admission practices and patterns of inpatient care for teens with AN. The interview questionnaire consisted of various items, including admission threshold for heart rate (HR) and percent ideal body weight (%IBW), two case vignettes followed by case-specific questions, and refeeding questions. Results: 51 out of 107 (48%) practitioners agreed to participate: 65% were female; 69% practiced in an academic setting; 16% in a group or solo practice; and 2% in a college health center. 84% were pediatricians and 88% had adolescent fellowship training. Participants represented 25 American states, one Canadian province, and 44 different adolescent programs. The HR thresholds for admission were: 50 (10%), 45 (23%), 40 (35%), 35 (17%), 30 (13%), or other (2%). The %IBW admission thresholds ranged from 60 – 85%: 85%IBW (2%), 75%IBW (52%), 70%IBW (38%), 65%IBW (5%), and 60%IBW (2%). There were no differences in admission practices based on number of years in practice, gender of practitioner, or practice setting. However, regional differences in admission practices existed with providers in the western United States less likely to admit patients with HR ⱖ 40 (p ⫽ 0.018). When presented with a case of a 12-year-old (y.o.) female with AN and a HR of 38, a weight 81% of IBW, and no recent weight change, 67% of respondents decided admission was warranted. Upon hearing a second case vignette, 60% of respondents decided they would admit a 16y.o. female with a decrease in weight from 76% to 73% of IBW and a HR of 49. In terms of refeeding practices, respondents described 28 different methods of advancing a diet during an admission. The forms of nutritional intake that practitioners usually chose varied: regular food (71%), oral liquid nutrition (55%), liquid nutrition via nasogastric (NG) feeding (37%), and intravenous fluids (12%). Only 37% of respondents
Abstracts / 36 (2005) 111-154
for example: increased fruit & vegetable consumption (1.6 to 2.6 servings/day, p ⫽ .03), decreased TV watching (2.8 to 2.3 hrs/day, p ⫽ .05), fewer pts eating fast food items ⱖ 3 days/wk (30% to 5%, p ⫽ .04). A higher percentage of involved pts lost wt (43%) than uninvolved patients (24%, p ⫽ .05) although, on average, pts from both groups gained wt (⫹0.24 lbs/mo vs. ⫹0.59 lbs/mo, NS) over a follow-up period that averaged 9 mos. Involved pts were significantly heavier than uninvolved pts (mean BMI 40.1 vs. 34.4 kg/m2). Pts who lost wt had significantly lower fasting insulin levels than those who did not. Conclusions: An intervention carried out in 3 SBHCs showed promise in modifying risk factors for type 2 DM among Hispanic and African-American high school students at high risk due to obesity. Support: Chicago Community Trust, Washington Square Health Foundation.
PII: S1054-139X(04)00384-2 81.
EMOTIONAL EATING IN ADOLESCENT FEMALES Cathleen Steinegger, M.D., Lorah Dorn, Ph.D., Cynthia Goody, Ph.D., R.D., L.D.#, Philip R. Khoury, M.S., and Stephen R. Daniels, M.D., Ph.D. Cincinnati Children’s Hospital Medical Center and the #University of Cincinnati Medical Center, Cincinnati, Ohio. Purpose: To investigate the emotional eating (EE) behavior of white and black adolescent females across puberty and adolescence. The overall aims include an examination of the patterns and predictors of EE, the relationship between the degree of EE and the average daily caloric intake, and whether EE influences final BMI. This report focuses on age and race-related patterns of EE in adolescents. Methods: This is a secondary data analysis of the National Heart, Lung and Blood Institute Growth and Health Study (NGHS) which collected data annually from 1987 through 1997 on black and white girls, beginning at 9 to 10 years of age. A subset of 870 girls, 431 white and 439 black, from one site was used in this analysis. Longitudinal data included: height, weight, pubertal stage (by exam), demographic data, 3-day food diaries, and the EmotionInduced Eating Scale (EIES) developed for the NGHS. The EIES is a 7 item scale with possible scores ranging from 7 (low EE) to 21 (high EE). Results: Using SAS®, a GEE model was used for analysis. The EE levels were low and, overall, age did not influence the EE score. However, at 9 –10 years, black girls had significantly higher levels of EE than white girls (10.4 vs. 9.8, p ⫽ 0.002). There was a significant age-by-race interaction with white girls increasing their EE score by 0.16 for each 1 year increase in age (p ⬍0.0001), while black girls decreased their EE score by 0.059 (p ⫽ 0.006) with each year. By study termination, 18 –19 years of age, the black girls’ were reporting lower EE scores than the white girls (9.5 vs. 10.2, p ⬍.001). Conclusions: The pattern of EE through adolescence varies by race. While black girls begin the adolescent years with higher EE levels than white girls, by 19 years of age, the races are reversed and white girls identify more EE behaviors than blacks. By identifying different patterns and influences on eating behavior between races, we may be able to create more effective nutrition interventions for disease prevention and treatment. Our subsequent analyses will explore the influence of puberty and pubertal timing,
depression, and socioeconomic status on the EE behavior followed by the role of EE on caloric intake and body mass.
PII: 1054-139X(04)00385-4 82.
OBESITY AND EATING DISORDERS IN OLDER ADOLESCENTS: DOES EARLY DIETING MAKE THINGS BETTER OR WORSE? Dianne Neumark-Sztainer, Ph.D., M.P.H., R.D. Melanie M. Wall, Ph.D. Marla E. Eisenberg, Sc.D., M.P.H. Mary Story Ph.D., R.D. University of Minnesota School of Public Health and School of Medicine. Minneapolis, Minnesota. Purpose: Eating and weight-related problems are of major public health concern owing to their high prevalence in adolescents and the potentially serious physical and psychosocial consequences. While dieting is commonly used as a weight loss strategy, we hypothesized that it may be exacerbating the problems of both obesity and eating disorders. The purpose of this study is to examine the prevalence of a range of weight-related problems in a large sample of older adolescents and to explore longitudinal associations between dieting in high school and weight gain, use of disordered eating behaviors, and the onset of eating disorders five years later in young adulthood. Methods: Project EAT was originally conducted in 1998 –1999 on an ethnically and socioeconomically diverse sample of middle school and high school students. In 2003–2004, participants were contacted and mailed surveys to assess eating and weight-related issues. This analysis focuses on the 1729 older adolescents who were followed from high school to young adulthood (Mean age ⫽ 20.4, SD ⫽ 0.89). Descriptive proportions are stratified by gender. Logistic regression is used to estimate odds ratios and 95% confidence intervals relating early dieting to later weight-related problems, adjusting for early weight-related problems. Results: Preliminary analyses show that 24% of the females and males were overweight or at-risk for overweight (BMI⬎ 85th percentile), using self-reported height and weight data. Binge eating was reported by 15.6% of females and 4.6% of males. Extreme weight control behaviors (vomiting, laxatives, diet pills, and diuretics) were reported by 24.3% of females and 6.6% of males. The presence of an eating disorder in the past year was reported by 5.7% of females and 0.5% of males. Preliminary analyses show that early dieting was significantly associated with becoming overweight in young adulthood (females: OR: 1.95, CI: 1.29,2.96; males: OR: 1.88, CI: 1.19,2.97) and with the later onset of binge eating (females: OR: 2.2; CI: 1.4,3.5; males: OR: 2.8; CI: 1.8, 6.0). Early dieting was associated with the onset of extreme weight control behaviors in females (OR: 2.2, CI: 1.6,3.2) and associations in males were of borderline significance (OR: 1.9; 1.0, 3.5). Early dieting was also associated with eating disorders in the past year in females (OR: 3.1, CI: 1.5,6.4) but associations in males were not statistically significant (OR: 2.3, CI: 0.4,14.0). Conclusions: In order to develop interventions that have the potential to prevent the broad spectrum of weight-related problems, we need to identify modifiable shared risk factors. Our findings suggest that dieting places youth at risk for eating disorders, disordered eating behaviors, and obesity. Efforts to prevent dieting, and, instead, encourage the adoption of long-term healthy
Abstracts / 36 (2005) 111-154
153
eating and physical activity behaviors are needed with clinics, schools, and in the broader community. Support: Grant # R40 MC 00319 – 02 from the Maternal and Child Health Bureau, Health Services Research Administration, U.S. Department of Health and Human Services (NeumarkSztainer, PI).
physiologic parameter may stabilize later than temperature and heart rate in adolescents recovering from malnutrition due to an eating disorder. Support: Partially supported by the Health and Resource Services Administration, Maternal and Child Health Bureau, Grant # 6 T71 MC00011.
PII: S1054-139X(04)00386-6
PII: S1054-139X(04)00388-X
83.
84.
INPATIENT TREATMENT OF ADOLESCENT PATIENTS WITH EATING DISORDERS: MEDICAL AND FINANCIAL OUTCOMES Jennifer L. Kalisvaart, M.D., Albert C. Hergenroeder, M.D., Section of Adolescent Medicine and Sports Medicine, Department of Pediatrics, Baylor College of Medicine, Houston, Texas.
VARIATIONS IN ADMISSION PRACTICES FOR ADOLESCENTS WITH ANOREXIA NERVOSA: A NORTH AMERICAN SAMPLE Jonathan Mansbach, MD, Beth Schwartz, BA, Jenna Gordon, MA, Debra K. Katzman, MD, and Sara Forman, MD. Divisions of Adolescent Medicine, Children’s Hospital Boston, Boston, MA, and The Hospital for Sick Children, Toronto, Ontario.
Purpose: To describe the medical outcomes and insurance reimbursement for treatment of patients with eating disorders on an adolescent medical unit. Methods: Using a descriptive, retrospective, cohort study design, we investigated the outcomes at the time of discharge for 39 patients with medical complications of anorexia nervosa (AN) or eating disorder not otherwise specified (EDNOS) following treatment on an adolescent medical unit, admitted over a three-year period. Admission criteria included a combination of estimated ideal body weight (%IBW) ⬍85%, bradycardia (resting heart rate ⱕ 50 bpm), hypothermia (ⱕ 35.5° C), and orthostatic tachycardia (increase ⱖ 35 bpm after standing for 5 minutes), and/or failure of outpatient treatment. 74% (N ⫽ 29) were diagnosed with anorexia nervosa (AN); 26% (N ⫽ 10) with eating disorder not otherwise specified (EDNOS). Outcomes included hospital and professional charges and reimbursements. Paired and independent samples ttests, Pearson Chi-square, and Wilcoxon rank sum analyses were used. Results: The mean age of the patients was 16.1 ⫾ 1.9 years. % IBW at admission was 74.8 %. Average daily weight gain was 0.100 ⫾ 0.06 kg/d. Hospital duration averaged 51 ⫾ 30 days. 64% were bradycardic and 44% were hypothermic at admission that improved to 17% and 0%, respectively, at discharge (p ⬍ .001 for both). Postural orthostatic tachycardia did not improve over the course of treatment. Discharge criteria included achieving 85% IBW. Medical diagnoses, including malnutrition, bradycardia, hypothermia and orthostatic tachycardia, were listed as the admitting and discharge diagnoses. Average total hospital and professional charges were $105,853 per patient, and insurance companies reimbursed an average of 62% of these total charges out of the patients’ medical benefits policies. No patient was discharged based on insurance criteria. Uniformly, the insurance companies denied coverage for the number of requested days. Conclusions: The majority of expenses for inpatient care on this adolescent medical unit were reimbursed by insurance companies under the patients’ medical benefits. However, the reimbursement was inadequate in order to achieve a discharge weight of 85% IBW, a weight associated with improved outcome. Inadequate reimbursement by insurance companies threatens the ability to provide optimal inpatient medical treatment for patients with eating disorders who are medically unstable. In addition, in contrast to what others have reported, normalization of orthostatic pulse may not be a criterion on which to base medical stabilization. This
Purpose: In 2000, the American Psychiatric Association published consensus admission guidelines for patients with eating disorders, but studies are needed to evaluate actual practices for clinical admission and treatment differences among patients with Anorexia Nervosa (AN). The purpose of this study was to assess the variability in admission practices and medical inpatient care for adolescent patients with AN. Methods: Members of the 2001–2004 Eating Disorder Special Interest Group from the Society for Adolescent Medicine were contacted via email and/or telephone. Participants completed a 30-minute structured telephone interview about their admission practices and patterns of inpatient care for teens with AN. The interview questionnaire consisted of various items, including admission threshold for heart rate (HR) and percent ideal body weight (%IBW), two case vignettes followed by case-specific questions, and refeeding questions. Results: 51 out of 107 (48%) practitioners agreed to participate: 65% were female; 69% practiced in an academic setting; 16% in a group or solo practice; and 2% in a college health center. 84% were pediatricians and 88% had adolescent fellowship training. Participants represented 25 American states, one Canadian province, and 44 different adolescent programs. The HR thresholds for admission were: 50 (10%), 45 (23%), 40 (35%), 35 (17%), 30 (13%), or other (2%). The %IBW admission thresholds ranged from 60 – 85%: 85%IBW (2%), 75%IBW (52%), 70%IBW (38%), 65%IBW (5%), and 60%IBW (2%). There were no differences in admission practices based on number of years in practice, gender of practitioner, or practice setting. However, regional differences in admission practices existed with providers in the western United States less likely to admit patients with HR ⱖ 40 (p ⫽ 0.018). When presented with a case of a 12-year-old (y.o.) female with AN and a HR of 38, a weight 81% of IBW, and no recent weight change, 67% of respondents decided admission was warranted. Upon hearing a second case vignette, 60% of respondents decided they would admit a 16y.o. female with a decrease in weight from 76% to 73% of IBW and a HR of 49. In terms of refeeding practices, respondents described 28 different methods of advancing a diet during an admission. The forms of nutritional intake that practitioners usually chose varied: regular food (71%), oral liquid nutrition (55%), liquid nutrition via nasogastric (NG) feeding (37%), and intravenous fluids (12%). Only 37% of respondents