 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
* Center for Demography and Ecology
Center for Demography of Health and Aging
Department of Population Health Sciences, University of Wisconsin, Madison, Wisconsin
|| Department of Sociology, University of Minnesota, Minneapolis, Minnesota
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Methods. A cross-sectional analysis was conducted using the 1996 National Longitudinal Study of Adolescent Health, a nationally representative sample of adolescents in grades 7 to 12 during the 1994–1995 school year, and 4743 adolescents with direct measures of height and weight. Using Centers for Disease Control and Prevention growth charts to determine percentiles, we used 5 body mass categories. Underweight was at or below the 5th percentile, normal BMI was between the 5th and 85th percentiles, at risk for overweight was between the 85th and 95th percentiles, overweight was between the 95th and 97th percentiles + 2 BMI units, and obese was at or above the 97th percentile + 2 BMI units. Four dimensions of health-related quality of life were measured: general health (self-reported general health), physical health (absence or presence of functional limitations and illness symptoms), emotional health (the Center for Epidemiologic Studies Depression Scale and Rosenberg's self-esteem scale), and a school and social functioning scale.
Results. We found a statistically significant relationship between BMI and general and physical health but not psychosocial outcomes. Adolescents who were overweight had significantly worse self-reported health (odds ratio [OR]: 2.17; 95% confidence interval [CI]: 1.34–3.51), as did obese adolescents (OR: 4.49; 95% CI: 2.87–7.03). Overweight (OR: 1.81; 95% CI: 1.22–2.68) and obese (OR: 1.91; 95% CI: 1.24–1.95) adolescents were also more likely to have a functional limitation. Only among the youngest adolescents (ages 12–14) did we find a significant deleterious impact of overweight and obesity on depression, self-esteem, and school/social functioning.
Conclusions. Using a nationally representative sample, we found that obesity in adolescence is linked with poor physical quality of life. However, in the general population, adolescents with above normal body mass did not report poorer emotional, school, or social functioning.
Key Words: adolescence • adolescent obesity • health-related quality of life • population-based studies
Abbreviations: HRQOL, health-related quality of life • Add Health, National Longitudinal Study of Adolescent Health • PedsQL, Pediatric Quality of Life Inventory • CESD, Center for Epidemiological Studies Depression Scale • OR, odds ratio • CI, confidence interval
The prevalence of childhood and adolescent overweight and obesity has increased substantially in the past 2 decades.1,2 Between 1986 and 1998, the prevalence of overweight and obesity increased among children and adolescents by 120% for blacks and Hispanics and by 50% for whites.1 Currently, 1 in 7 children and adolescents in the United States is overweight. Given this staggering increase, researchers, physicians, and parents have become increasingly concerned about both the short- and long-term health and psychosocial consequences of childhood and adolescent obesity.
Previous research has attempted to asses the impact of overweight on children's physical and psychosocial outcomes. Traditionally, these studies examined 1 outcome (eg, self-esteem, glucose intolerance) in isolation from others.3–11 Previous research on the relationship between childhood overweight and psychosocial outcomes has been inconsistent, finding either no association or a small inverse relationship.3,4,11 The association between overweight and physical health has been much more consistent. Overweight has been linked to multiple physical health problems in children and adolescents, including glucose intolerance and risk factors for cardiovascular disease.7–11
Recently, researchers have begun to document the impact of elevated BMI on health-related quality of life (HRQOL). HRQOL is a construct that attempts to provide a generalized assessment of well-being measured along multiple dimensions, including physical, functional, psychological, and social well-being.12 One recent study found serious adverse consequences of obesity on HRQOL in a clinical sample of severely obese (mean BMI: 34.7) children and adolescents 5 to 18 years of age.13
Although clinical samples are frequently the first to identify important associations, they are often subject to biasing effects of nonrandom sample selection and can be too small to test for potential confounding effects. Population-based studies have also examined the association between overweight and HRQOL; the association between overweight and psychosocial HRQOL has been less consistent in population-based studies than in clinical samples. Previous population-based studies have also been limited by small sample size, lack of adequate control for family sociodemographic characteristics, and an inadequate focus on adolescents.14,15
Recent work has suggested that the effect of body mass on well-being may indeed vary by sociodemographic characteristics. In particular, the psychosocial elements of HRQOL may be differentially affected by gender, race/ethnicity, and age. Although there are no gender differences in obesity rates in children and adolescents,2 the stigma associated with obesity is thought to be greater for girls than for boys.16,17 Researchers have also found that overweight girls are more depressed and have lower self-esteem than normal-weight girls but that overweight boys are similar to normal-weight boys on these outcomes.3,4
Others have documented that BMI varies by race and ethnicity,18 and there are important racial/ethnic differences in the relationship between BMI and psychosocial outcomes. For example, white female adolescents are more likely than black female adolescents to view themselves as overweight, engage in unhealthy weight management behaviors, and have low self-esteem.4,19,20
Finally, previous research suggests that the adverse psychosocial consequences of overweight and obesity are strongly related to age and developmental status.14 For example, obesity was not associated with psychosocial outcomes in children,5 but during adolescence, it became a stronger predictor of poor psychosocial outcomes.21
We analyzed the relationship between BMI and various measures of HRQOL using the National Longitudinal Study of Adolescent Health (Add Health), a nationally representative sample of adolescents who were in grades 7 to 12 during the 1994–1995 school year. This analysis complements earlier work on primary school–aged children and also allows for the control of important sociodemographic and family characteristics not controlled in previous analyses.14,15 In addition, because previous research has suggested that the relationship between BMI and HRQOL is j-shaped, this study investigates the effect of both underweight and overweight on HRQOL.15–22
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Measures: Predictor Variables
BMI
BMI is a widely used measure of adiposity that is calculated as weight in kilograms divided by height in meters, squared (kg/m2). Because adolescent girls in particular underreport their weight,24 one key advantage of Wave 2 of Add Health was the inclusion of direct anthropometric measures of height and weight. Including only those adolescents with directly measured height and weight yielded a sample of 4743 adolescents; 84 (1.7%) of the total sample were excluded for missing height or weight.
When assessed within particular age and gender groups, BMI is a statistically valid measure of overweight among children and adolescents.25 We used growth charts provided by the Centers for Disease Control and Prevention to determine BMI percentiles for boys and girls of each age.26 Previous research has defined overweight as at or above the 95th percentile and "at risk" for overweight as at or above the 85th percentile but below the 95th percentile.2 We used 5 mutually exclusive categories. Underweight is at or below the 5th percentile,27,28 normal BMI is between the 5th and 85th percentiles, at risk for overweight is between the 85th and 95th percentiles, overweight is between the 95th and 97th percentiles plus 2 BMI units, and obese is at or above the 97th percentile plus 2 BMI units. (We added 2 BMI units to the 97th percentile to achieve a reasonably balanced distribution of overweight and obese adolescents.) Table 1 provides the descriptive statistics for Add Health respondents with BMI information available. Mean BMI in the Add Health sample was 
23.
|
We also included measures of family income. Because roughly one fifth of the cases had missing data for family income, we imputed family income for those cases as a linear function of parental education and family structure. Retaining these adolescents preserved statistical power and protected against biasing the sample. Imputed values were combined with reported values and categorized into <$20000, $20000 to $44999, $45000 to $74999, and $75000 or more. In addition, because adolescents with missing data on income may possess qualities that were not accounted for by our set of control variables (eg, wealth), we created a dummy variable for imputed cases to determine whether they were different from other adolescents in terms of HRQOL.
Measures: Outcome Variables
Ideally, we would prefer a broad measure of HRQOL, such as the Pediatric Quality of Life Inventory (PedsQL),29 which was used to assess the HRQOL of children and adolescents in a recent clinical sample.13 All 23 items in the PedsQL provided a general measure of HRQOL. The PedsQL can be separated into the following dimensions: physical health, emotional functioning, social functioning, and school functioning. We approximated PedsQL by using 1 measure of general health, 2 measures of physical health (functional limitations and illness symptoms), 2 measures of emotional functioning (depression and self-esteem), and 1 combined measure of school and social functioning. With the exception of general health and functional limitations, all of our outcomes were dichotomized using a distributional approach that identified adolescents who were at least 1 SD away from the mean (toward the poor functioning tail of the distribution). This threshold was chosen not only because it identified adolescents who were substantially different from the norm but also for comparability with previous research.13 We tested ordinary least square regression models with continuous outcomes as well and found no substantial differences between these models and the models with dichotomized outcomes in terms of the relationship between BMI and HRQOL. Table 2 provides descriptive statistics for the 6 outcome measures among Add Health respondents with BMI information available.
|
Physical Health
Wave 2 included a set of questions on functional limitations (limitations attending school, difficulty performing household chores, limitations doing strenuous acts, and difficulty with personal care and hygiene). Respondents who answered "yes" to any of these questions were categorized as "limited."
Respondents also were asked how many times each of 13 illness symptoms were experienced in the past year, and an index with possible values ranging from 0 to 52 was created. A high level of illness symptoms was defined as 1 SD or more above the mean (ie, 15 or more symptoms).
Emotional Health
Emotional health was assessed using items from the Center for Epidemiologic Studies Depression Scale (CESD)30 and Rosenberg's self-esteem scale.31 Nineteen items from the CESD were used to assess depression.32 The CESD has been previously validated in adolescents33,34 and adults.30 An index with possible values ranging from 0 to 57 was constructed and then dichotomized at 19.
Six items from Rosenberg's self-esteem scale31 included in Add Health asked respondents whether they have good qualities, have a lot to be proud of, like themselves as they are, always do things right, feel socially accepted, and feel loved and wanted. A reverse-coded index with possible scores ranging from 0 to 24 was created and dichotomized at 9. Higher scores indicate lower self-esteem.
School and Social Functioning
Insufficient items on school and social functioning were present in Add Health to permit the construction of separate indexes for these domains. Therefore, a single index that consisted of 9 items was created. Four items asked whether the adolescent had trouble getting along with teachers, getting along with other students, paying attention, and getting homework done. Another 4 items asked respondents whether they felt close to people at school, part of their school, safe at school, and happy at school. The final item asked how often the respondent had missed school with an excuse in the 1995–1996 school year. The potential range of the index was 0 to 36, and it was dichotomized at 16.
To determine how reliably our school and sociability items measure a single underlying construct, we conducted a Cronbach's 
analysis. This yielded a reliability coefficient of .73, which indicates that the school and sociability items are generally reliable measures of the same construct. There are fewer cases with scores (300 cases missing; 6.3%) on this index than for the other outcome variables because the Add Health survey did not ask questions about school of students who were interviewed during the academic year but were not attending school for some reason or students who were interviewed in the summer and reported that they had not attended school for the entirety of the previous academic year.
Data Analyses
SAS for Windows, version 8.02, was used to manage data and generate descriptive statistics.35 The effect of BMI (and other predictors) on binary measures of HRQOL was estimated via the SVYLOGIT procedure in Stata/SE, version 8.1, which uses a pseudo-likelihood estimator to fit logistic regression models because traditional estimation techniques (eg, Fisher scoring) do not produce a true likelihood under complex survey designs.36 SVYLOGIT enabled us to correct for both sampling probabilities and clustering in the primary sampling units. Accounting for survey design effects permitted us to produce unbiased estimates of standard errors. We tested simple models with only BMI, full models with potential mediators and confounders, and stratified models, whereby we examined whether the effect of BMI on HRQOL outcomes was modified by age, gender, or race.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Of the potential confounding factors considered (see Table 4), gender seemed the most important. Gender was a significant predictor of every outcome. Girls were significantly more likely than boys to report poor general health, functional limitations, many illness symptoms, depression, and low self-esteem. Girls were more likely than boys to report high school/social functioning.
|
In the full models (Table 4), age generally was not an independent predictor of HRQOL for our sample, although younger (ages 12–14) adolescents were less likely than adolescents who were aged 15 to 17 to be depressed or report low self-esteem. Also, the oldest adolescents (ages 18–20) were less likely than adolescents who were aged 15 to 17 to report many illness symptoms. Because there were some indications14,21 that obesity may affect HRQOL in an age-specific manner, we also tested the impact of obesity among 3 groups of adolescents. Surprising, we found that only in the youngest group of adolescents (ages 12–14) did body mass exert any particular effect on psychosocial HRQOL. The 12- to 14-year-olds were significantly more likely to be depressed when overweight (OR: 3.04; 95% CI: 1.19–7.76) or obese (OR: 2.83; 95% CI: 1.25–6.41), controlling for all other confounders. The obese 12- to 14-year-olds were also significantly more likely to report low self-esteem (OR: 3.47; 95% CI: 1.30–9.24) and poor school/social functioning (OR: 2.33; 95% CI: 1.15, 4.72) compared with 12- to 14-year-olds with normal BMI.
Race was only occasionally important in predicting HRQOL in our full models and very heterogeneous in its effect (Table 4). Hispanics were more likely than whites to report poor general health, depression, and low self-esteem. Asians also were more likely than whites to report depression and low self-esteem. Blacks, however, were much less likely than whites to report low self-esteem (OR: 0.44; 95% CI: 0.32–0.62). Finally, adolescents in the "other race" category were more likely than whites to report many illness symptoms.
We also tested whether BMI may operate differently for blacks than for whites in race-specific models. The 1 notable difference that we found was in functional limitations; for blacks, body mass was not a significant predictor of functional limitations (underweight OR: 3.05 [95% CI: 0.65–14.41]; at risk for overweight OR: 0.69 [95% CI: 0.36–1.35]; overweight OR: 1.55 [95% CI: 0.60–4.04]; obese OR: 1.29 [95% CI: 0.47–3.56]). In contrast, compared with whites with normal BMI, every nonnormal body mass group among whites had significantly elevated functional limitations (underweight OR: 2.07 [95% CI: 1.00–4.27]; at risk for overweight OR: 1.48 [95% CI: 1.00–2.18]; overweight OR: 2.13 [95% CI: 1.33–3.41]; obese OR: 2.71 [95% CI: 1.67–4.39]).
Family structure was generally not important for the physical outcomes but was a significant predictor of many of the psychosocial outcomes (Table 4). Adolescents who resided in single-parent, step-parent, or "other" type family were more likely to be depressed, have low self-esteem, and have poor school functioning relative to their peers in 2-parent families.
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This study has several unique strengths. First, we tested the link between BMI and adolescent HRQOL with a nationally representative sample. This allows greater generalizability than past studies that examined similar issues. Second, the sociodemographic heterogeneity of our sample allowed us to examine differential effects of BMI by gender, race, age, and socioeconomic status. This revealed some interesting nuances in the relationship between BMI and adolescent HRQOL. Third, we were able to capture each of the domains included in standard measures of HRQOL, such as PedsQL, by creating or using separate measures for each domain. Typically, secondary data from large national surveys are limited in the availability of measures in multiple domains of interest. Fortunately, the Add Health data set contains several measures in each of the HRQOL domains. The strengths of this study complement existing studies and can significantly enhance the current literature, which is based largely on clinical and community-based samples.13–15
Despite these strengths, we note a number of limitations. First, Add Health is a school-based sample and thus excludes adolescents who are not in school. Because school is compulsory until age 16, only adolescents in the older age groups are likely to be missed. This could disproportionately select out overweight or obese adolescents, although we know of no evidence on differential school enrollments by weight.
Second, because a direct measure of BMI and several of our outcome measures were available only in Wave 2, we were not able to assess the BMI–HRQOL relationship longitudinally. We note that most current research seems to agree that elevated body mass tends to cause lower HRQOL rather than low HRQOL causing body mass elevation,37 but we cannot demonstrate causality without longitudinal data.
Third, independently validated measures were not available for all of our outcomes. We used the best measures available and attempted to stay true to the variables included in the PedsQL. In particular, we note that depression is a high standard for emotional health and may not capture more subtle declines in emotional health. Our measure of school and social functioning is also imperfect because it does not include information about teasing.38
Fourth, although currently the best data for our question, the Add Health data used here are already almost 8 years old. As newer population-based samples of adolescents become available, they could further extend our understanding of the current HRQOL of overweight and obese adolescents, particularly in regard to psychosocial outcomes.
Finally, although the number of cases in our data set is large, in analyses stratified by sociodemographic measures (age group, race, etc), the subgroups become substantially smaller with a resultant loss of power. Thus, we may not be able to detect statistical significance for effects that are large. If this is true, then our estimates of the BMI–HRQOL relationship by some sociodemographic characteristics may be underestimated.
Relative to other studies on the adolescent BMI–HRQOL relationship, our study's strengths more than compensate for its weaknesses. For example, although our study excludes those who are not in school, other studies exclude those who do not seek clinical treatment.13 In general, we know people attend medical clinics because they believe that they have symptoms indicating a health problem.39–43 Previous research has indicated that obese adults who sought treatment had significantly worse HRQOL than obese adults who did not seek treatment.39,40,44 Obese adolescents who seek treatment similarly may have worse HRQOL than obese adolescents who do not seek treatment.
This study has raised several questions for future research. First, we find several differential effects by sociodemographic characteristics: (1) body mass is more strongly associated with functional limitations for boys than for girls, (2) nonnormal BMI is associated with functional limitations for whites but not for blacks, and (3) BMI is associated with poor social and emotional functioning only for those who are 14 years of age or younger. These findings raise questions about the mechanisms that generate these differences by gender, race, and age.
Finally, our findings beg the question, "Why is there not a relationship between BMI and psychosocial HRQOL for most adolescents?" Perhaps the new cohort of young Americans is more tolerant of weight differences than previous cohorts. Previous research has demonstrated more tolerance of political and sexual nonconformity among younger Americans,45,46 suggesting that BMI differences in general and obesity in particular may also be more acceptable. Perhaps it is simply that being overweight is now more common among adolescents.1
Alternatively, it is possible that adolescents underreport their health problems. Although there is some literature on children's reports of HRQOL, the findings are mixed. Comparing child and parental proxy or clinician reports, some studies indicate that adolescents are much less optimistic about their health and well-being than their parents,47,48 whereas other studies suggest that children with health conditions are more optimistic than is warranted given their condition.49 Some studies also find that, compared with parents, children's and adolescents' reports more closely match clinicians' assessments.50 Given this mixed evidence, it is not clear whether the lack of an association between BMI and psychosocial HRQOL in the general adolescent population is attributable to systematic misreporting. Add Health data do not allow for investigation of most of these questions, but we are hopeful that future data collection efforts will focus on mechanisms that may help to explain our differential findings by gender, race, and age and the lack of a significant relationship between body mass and psychosocial HRQOL among most adolescents.
|
ACKNOWLEDGMENTS |
|---|
We thank Larry Bumpass for drawing our attention to the initial reports of the impact of obesity on quality of life.
|
FOOTNOTES |
|---|
Reprint requests to (K.C.S.) Center for Demography of Health and Aging, University of Wisconsin, 4424 Social Science Bldg, 1180 Observatory Dr, Madison, WI 53705. E-mail: kswallen{at}ssc.wisc.edu
No conflict of interest declared.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. K. Kumanyika, E. Obarzanek, N. Stettler, R. Bell, A. E. Field, S. P. Fortmann, B. A. Franklin, M. W. Gillman, C. E. Lewis, W. C. Poston II, et al. Population-Based Prevention of Obesity: The Need for Comprehensive Promotion of Healthful Eating, Physical Activity, and Energy Balance: A Scientific Statement From American Heart Association Council on Epidemiology and Prevention, Interdisciplinary Committee for Prevention (Formerly the Expert Panel on Population and Prevention Science) Circulation, July 22, 2008; 118(4): 428 - 464. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. C. Modi, S. M. Guilfoyle, and M. H. Zeller Impaired Health-related Quality of Life in Caregivers of Youth Seeking Obesity Treatment J. Pediatr. Psychol., July 18, 2008; (2008) jsn070v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. E. Waring and K. L. Lapane Overweight in Children and Adolescents in Relation to Attention-Deficit/Hyperactivity Disorder: Results From a National Sample Pediatrics, July 1, 2008; 122(1): e1 - e6. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-A. Carno, E. Ellis, E. Anson, R. Kraus, J. Black, R. Short, and H. V. Connolly Symptoms of Sleep Apnea and Polysomnography as Predictors of Poor Quality of Life in Overweight Children and Adolescents J. Pediatr. Psychol., April 1, 2008; 33(3): 269 - 278. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. C. Skinner, M. L. Mayer, K. Flower, and M. Weinberger Health Status and Health Care Expenditures in a Nationally Representative Sample: How Do Overweight and Healthy-Weight Children Compare? Pediatrics, February 1, 2008; 121(2): e269 - e277. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. R. Lemeshow, L. Fisher, E. Goodman, I. Kawachi, C. S. Berkey, and G. A. Colditz Subjective Social Status in the School and Change in Adiposity in Female Adolescents: Findings From a Prospective Cohort Study Arch Pediatr Adolesc Med, January 1, 2008; 162(1): 23 - 28. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. E. Simon, K. S. Chan, and C. B. Forrest Assessment of Children's Health-Related Quality of Life in the United States With a Multidimensional Index Pediatrics, January 1, 2008; 121(1): e118 - e126. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. W. Lowry, B. J. Sallinen, and D. M. Janicke The Effects of Weight Management Programs on Self-Esteem in Pediatric Overweight Populations J. Pediatr. Psychol., November 1, 2007; 32(10): 1179 - 1195. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. F. Wilson Adolescents' Attitudes About Obesity and What They Want in Obesity Prevention Programs The Journal of School Nursing, August 1, 2007; 23(4): 229 - 238. [Abstract] [PDF]
|
|
|
|
|
|
G. S. Goldfield, R. Mallory, T. Parker, T. Cunningham, C. Legg, A. Lumb, K. Parker, D. Prud'homme, and K. B. Adamo Effects of Modifying Physical Activity and Sedentary Behavior on Psychosocial Adjustment in Overweight/Obese Children J. Pediatr. Psychol., August 1, 2007; 32(7): 783 - 793. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. R. Piquero, L. E. Daigle, C. Gibson, N. L. Piquero, and S. G. Tibbetts Research Note: Are Life-Course-Persistent Offenders At Risk for Adverse Health Outcomes? Journal of Research in Crime and Delinquency, May 1, 2007; 44(2): 185 - 207. [Abstract] [PDF]
|
|
|
|
|
|
D. W. Beebe, D. Lewin, M. Zeller, M. McCabe, K. MacLeod, S. R. Daniels, and R. Amin Sleep in Overweight Adolescents: Shorter Sleep, Poorer Sleep Quality, Sleepiness, and Sleep-Disordered Breathing J. Pediatr. Psychol., January 1, 2007; 32(1): 69 - 79. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Elwing, P. J. Lustman, H. L. Wang, and R. E. Clouse Depression, Anxiety, and Nonalcoholic Steatohepatitis Psychosom Med, July 1, 2006; 68(4): 563 - 569. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Hahn, S. Benson, S. Elsenbruch, K. Pleger, S. Tan, K. Mann, M. Schedlowski, W. B. van Halteren, R. Kimmig, and O. E. Janssen Metformin treatment of polycystic ovary syndrome improves health-related quality-of-life, emotional distress and sexuality Hum. Reprod., July 1, 2006; 21(7): 1925 - 1934. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. O'Lonergan and J. J. Zodrow Pediatric Assent: Subject Protection Issues among Adolescent Females Enrolled in Research J. Law Med. Ethics, June 1, 2006; 34(2): 451 - 459. [PDF]
|
|
|
|
 |
|
 S. R.S. Stender, G. A. Burghen, and J. T. Mallare The Role of Health Care Providers in the Prevention of Overweight and Type 2 Diabetes in Children and Adolescents Diabetes Spectr, October 1, 2005; 18(4): 240 - 248. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
