 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
* Pediatric Psychopharmacology Unit of the Psychiatry Department, Massachusetts General Hospital
Harvard Medical School, Boston, Massachusetts
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Methods. Height and weight were examined in 124 female ADHD children and 116 female controls using age and parental height corrections, attending to issues of pubertal stage and treatment. Also, we examined the interaction between ADHD status and gender on growth outcomes using data from 124 ADHD and 109 control males.
Results. The ADHD-growth association was not moderated by gender. No deficits in age-adjusted height or age and height-adjusted weight were detected in ADHD girls. Also, we found no association between growth measurements and psychotropic treatment, malnutrition, short stature, pubertal development, family history of ADHD, or psychiatric comorbidity, except for major depression: ADHD girls with major depression were on average 7.6 kg heavier than ADHD girls without depression, adjusting for age and height.
Conclusions. No growth deficits appear to be associated with ADHD or its treatment in females. These findings add to a growing literature supporting the notion that stimulant treatment does not have an adverse impact on ADHD children’s growth and development.
Key Words: ADHD • growth • stimulants
Abbreviations: ADHD, attention-deficit/hyperactivity disorder • DSM-III-R, Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Revised • MD, major depression • SES, socioeconomic status
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
In contrast to these findings, Satterfield et al7 reported that initial height deficits observed in ADHD children after 1 year of stimulant treatment dissipated by the second year of treatment despite persistent weight deficits and uninterrupted treatment. A more recent study by our group assessed this issue using several methodological advances in the assessment of height deficits, including corrections by parental height and a conversion to z scores that corrects for age, avoids mathematical distortions, and is sensitive to height changes at all levels of height.8 In this study, Spencer et al9 reported small but significant differences in height in ADHD youth compared with those without this disorder, but showed that these height deficits were evident in early but not late adolescent ADHD children and were unrelated to use of psychotropic medications. Taken together, these findings raise the possibility that height deficits in ADHD children may reflect temporary developmental deviations associated with ADHD, not complications of stimulant treatment.
Although reassuring, a major shortcoming of the extant literature assessing growth deficits in ADHD is the exclusive limitation to samples of boys. Recent work by us and others clearly documents that ADHD is also highly prevalent in girls, similarly familial and highly comorbid with other disruptive behavior, mood, and anxiety disorders and that females with ADHD are likely to receive treatment with stimulant medications.10–21 Thus, additional information is needed to determine if girls with ADHD exposed to stimulant medication are vulnerable to growth deficits. Also, it is possible that the relationship between ADHD and growth may be modified by gender, an issue that has not been previously examined.
The purpose of this report was the systematic evaluation of growth deficits in girls with ADHD attending to issues of gender, therapeutics, comorbidity, and familiality using developmentally sensitive methodology. Based on the previous analysis in boys with ADHD, we expected to find small growth deficits in height in ADHD girls that occur prepubertally, and that these deficits will be independent of stimulant treatment. To our knowledge, this work represents the first evaluation of growth deficits in ADHD girls.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
A 3-stage ascertainment procedure was used to select the subjects. For ADHD subjects, the first stage was the patient’s referral. The second stage confirmed the diagnosis of ADHD by using a telephone questionnaire administered to the mother. The questionnaire asked about the 14 DSM-III-R symptoms of ADHD and questions regarding study exclusion criteria. The third stage confirmed the diagnosis with face-to-face structured interviews with the mother. Only patients who received a positive diagnosis at all 3 stages were included.
Control proband selection was guided by contemporary epidemiologic methodology, which dictates that the sampling of controls should be drawn from the exposure distribution of the source population that gave rise to the cases.22 As this was a clinic-based case series, the source population can be defined as the group of children who would have received treatment for ADHD at the same clinics as the cases if they, contrary to fact, had ADHD. In addition, this control selection should be conducted independent of exposure, with the only criteria for inclusion being the absence of ADHD. Thus, we ascertained control participants from referrals for routine physical examinations to the same medical clinics that provided the case series. In stage 2, the control mothers responded to the telephone questionnaire. Eligible controls meeting study entry criteria were recruited for the study and received the third stage assessment (structured interview). Only subjects classified as not having ADHD at all three stages were included in the control group. As the primary aim of the study was the examination of the familial risk of ADHD, cases and controls were not matched on any potentially confounding variables. In the present study, confounders are dealt with in the analysis phase.
Diagnostic Assessments
We used DSM-III-R-based structured interviews to diagnose subjects supplemented with questions that would allow us to make DSM-IV diagnoses. Psychiatric assessments of probands were made with the Kiddie SADS-E (Epidemiologic Version, supplied by H. Orvaschel and J. Puig-Antich, Nova Southeastern University, Fort Lauderdale, FL).23 Diagnoses were based on independent interviews with the mothers and direct interviews with the child. Children younger than 12 years of age were not interviewed directly. Kappa coefficients of agreement were computed between raters and 3 board-certified psychiatrists who listened to audiotaped interviews. Based on 173 interviews, the median kappa was .86 and the kappa for ADHD was .99. We assessed socioeconomic status (SES) with the Hollingshead-Redlich scale24 and functioning with Global Assessment of Functioning scale.
A sign-off committee of board-certified child and adult psychiatrists chaired by the Program Director (J.B.) resolved all diagnostic uncertainties. As suggested by others,25,26 we diagnosed major depression (MD) only if the depressive episode was associated with marked impairment. Since the anxiety disorders comprise many syndromes with a wide range of severity, we used 2 or more anxiety disorders to indicate the presence of a clinically meaningful anxiety syndrome and refer to this as "multiple anxiety disorders" as we have elsewhere.27
Assessments of Growth and Pubertal Development
All probands and relatives were weighed and measured using the same scale. Measurements were obtained with the subjects lightly clothed but without shoes. We used a Physician’s Beam Scale (Detecto, Webb City, MO), a high calibration scale with a height rod for precise measurement of height. Subjects were erect with height examined at the vertex. Growth measurements were plotted on National Center for Health Statistics growth tables.28 These growth charts are sex-specific and standardized. Thus, they permit comparisons of growth deficit findings to normal population data.
To assess pubertal staging, children 12 to 18 years of age were asked questions about pubertal development. Questions included the presence of pubertal hair, axillary hair, and menses as well as the age at attainment of each stage. Based on these questions, estimates of Tanner stages were developed as follows: attainment of pubertal hair, Tanner stage 2 to 3; attainment of axillary hair, Tanner stage 3 to 4; menstruation, Tanner stage 4 to 5.
Based on growth measurements, the following height and weight indices were made, as in our investigation with boys with ADHD:9
Data Analysis
First, we tested the difference between ADHD and control females on demographic features. Age and ages of onset were compared with t tests. Rates of pharmacotherapy between ADHD and controls were analyzed by the 
2 test, and SES by the rank sum test. Then, to assess the role of gender as a potential modifier of the relationship between ADHD and growth, we modeled growth outcomes (metric height and weight as well as the adjusted measures described above) as a function of ADHD status, gender, and their interaction using linear regression. Further analyses of growth outcomes were also analyzed using linear regression. Associations between height and weight measurements were examined using the Pearson correlation. All analyses were 2-tailed and statistical significance was defined at the .05 level.
To calculate power for the comparison of ADHD to control girls on height z scores, we used Cohen’s effect size scale40 to determine the power for medium-sized effects. Using an estimated standard deviation of 1.2 and an 
level of .05, we calculated our power to detect medium-sized effects of .4 and .5, corresponding to mean height z score differences of .48 and .60, to be 87% and 97%, respectively. These mean z score differences of .48 and .60 correspond to differences of 18 and 23 percentile points, respectively. A 25-percentile point difference (from the 50th percentile to the 25th percentile) translates into height differences of 1.3 inches to 1.9 inches, depending on age. Thus, we have adequate power to detect the minimum of clinically meaningful height effects.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Growth in Height
Control probands were significantly taller (mean ± 6 cm, P = .005) than ADHD girls (Table 2). However, when heights were converted to age-specific z scores, the difference in height between ADHD and controls was no longer significantly different. Similar results were obtained after correcting by parental height and proband age. The same pattern of results was found when the analysis was restricted to children (age <12 years) or adolescents (age 
12 years). There were no statistically or clinically meaningful differences in height between parents of ADHD and control children. Girls’ heights were moderately correlated with mothers’ heights in both ADHD (r = 0.22, P = .017) and control groups (r = .23, P = .014). However, correlations with fathers’ heights were much weaker, in both ADHD (r = -.05, P = .607) and control groups (r = .09, P = .359).
|
Impact of Psychopharmacologic Treatment on Growth
No meaningful differences in any height measurement were detected between psychopharmacologically treated (either lifetime or in the past 2 years) and untreated ADHD subjects (Table 3). Surprisingly, a statistically significant difference was found in absolute weight, with medicated ADHD girls being heavier than unmedicated ones (mean difference 7 kg; P < .05). Moreover, medicated ADHD girls were consistently taller and heavier than their nontreated counterparts on every height and weight measure examined.
|
Impact of Psychiatric Comorbidity and Family History on Growth
Measures of height and weight did not significantly differ between ADHD girls with and without conduct disorder, multiple anxiety disorders, or a family history of ADHD (all P values >.05). However, ADHD girls with comorbid MD had a significantly greater average height- and age-corrected weight index relative to ADHD girls without MD (with MD, N = 21: 126 ± 31.3; without MD, N = 103: 109 ± 25.7; t122 = -2.6, P = .011). To enhance the interpretability of this finding, we modeled weight as a function of MD status, height, and age in a linear regression model. We found that MD females were on average 7.6 kg (16.7 pounds) heavier than female children without MD, holding height and age constant. It should be noted that the average weight index of the ADHD children with MD was greater than the recommended cutoff for obesity, 120.32,36
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our finding of a modest numerical height deficit in preadolescent ADHD proband girls compared with controls is consistent with our previous findings in boys. That work9 showed that preadolescent boys with ADHD were significantly smaller than boys without ADHD of the same age but that this effect was not apparent in older probands. However, in contrast to the findings in boys, the small height difference in girl probands disappeared after corrections by age and parental height. Although this differential association between ADHD and age- and parental-corrected height was not detected by the gender interaction model, these findings provide preliminary evidence that gender may moderate the association between ADHD and delays in the tempo of growth in height with males being slightly more affected than females. As interaction effects are hampered by low statistical power, future studies with adequate sample sizes should reexamine this gender effect to further clarify this issue.
Also, as previously documented in boys, we failed to find evidence for meaningful effects of stimulant treatment on growth in height in our sample of ADHD girls. This finding extends to girls previously documented findings in boys indicating that stimulant treatment does not detrimentally impact growth in height in children with ADHD. In contrast to our results, 3 of the 6 previous studies that compared height deficits in treated and untreated ADHD children2,3,41 reported stimulant-associated height deficits.42–44 However, these studies evaluated preadolescent samples and could not fully assess normalization of height over time. Similarly, although 3 of 4 studies that evaluated the impact of drug holidays in ADHD children found that continued stimulant treatment was associated with height suppression and that rebound growth occurred during drug holidays,1,3,7,45 it remains unclear whether increased growth in height during drug holidays reflects spontaneous normalization of growth in height over time.7,46,47 Moreover, in contrast to previous studies examining height in ADHD youth that used >8 different methods of assessing growth, including direct comparisons of averaged absolute height or percentiles from standardized growth charts,31 methods subject to artifactual distortion and low sensitivity, our study benefited from the use of z scores to assess height deficit.8 Since z scores are more sensitive and less vulnerable to distortions in a sample with a wide range of ages such as ours, they more accurately reflect ADHD-associated height deficits than absolute height. Despite these issues, it seems prudent that children suspected of significant stimulant-associated growth deficits be provided with drug holidays or alternative treatment to help mitigate this problem until further research can confirm our findings. This recommendation should be carefully weighed against the risk for exacerbation of symptoms attributable to drug discontinuation.
As previously documented in boys,9 no meaningful associations were identified between ADHD or its treatment and growth in weight in females with ADHD. In fact, there was evidence of more than adequate body mass in our sample. Thus, these results contradict findings from an early literature that suggested a detrimental impact of stimulant treatment on growth in weight.2,4,5,7,31,43,48
Although an early report linked stimulant-associated weight loss with height loss,2 more recent reports, including ours, have failed to replicate this finding.1,7,31,45,46 These findings add to the growing consensus that stimulant-associated weight deficits are not implicated in ADHD-associated height deficits.
As previously reported in boys,9 no meaningful associations were identified between ADHD or its treatment and pubertal development (Tanner stages) in this large sample of girls with and without ADHD. The ages of onset of Tanner stages were equivalent between the groups in both genders and consistent with the published ages of onset of Tanner stages in the general population.49,50 Taken together, these results provide strong support for the notion that neither ADHD nor its treatment influences pubertal development.
Interestingly, a noteworthy association was found between comorbidity with depression and significant weight gain in females with ADHD. Since no such association was previously observed in male probands with ADHD, these results can be interpreted as suggesting that depression may put female youth with ADHD at high risk to overeat and excessive weight gain. Alternatively, being overweight or obese may place ADHD girls at risk for depression. Also, there could be a third, unmeasured factor acting in ADHD females that contributes to the occurrence of both depression and overweight, such as a metabolic disturbance. Although our cross-sectional study does not allow us to explore these issues, future research efforts using longitudinal designs should attend to them.
The findings reported here should be evaluated against the methodologic limitations. Pubertal stages were based on self-report, not on direct physical examination. However, studies have shown that self-assessed pubertal stages may be highly concordant with physician-assessed pubertal staging.51 Since ours was a cross-sectional study, it permits only weak developmental inferences. However, we could compare growth parameters in younger and older subjects, generating initial developmental hypotheses to be tested in future longitudinal studies. The subgroup analyses comparing ADHD probands stratified by medication status were not as well-powered as our primary comparison of ADHD and control probands. However, we still had 80% power to detect the medium-sized effect of .5, which translates into 23 percentile points, or a height difference of 
1.2 to 1.8 inches, depending on age. Thus, although not as powerful as the primary analyses, the subgroup tests still were powerful enough to detect medium-sized and clinically relevant effects. Finally, our analyses of ADHD girls stratified by stimulant treatment would have benefited from additional information regarding the duration, dose, and interruptions of treatment. Since this was an observational study and not a clinical trial, we do not have detailed information on these parameters.
Despite these limitations, our findings demonstrate minimal deficits in growth in height in preadolescent ADHD girls unrelated to pharmacological treatments. We found no evidence of delayed pubertal development, weight deficits, or a relationship between measures of malnutrition and short stature. These findings extend to females previous findings reported in boys documenting that neither ADHD nor its treatment are associated with a detrimental effect on growth in height or weight. These cross-sectional findings should begin to offset prevailing concerns regarding putative detrimental effects of the pharmacotherapy of ADHD on development and growth. Additional, longitudinal studies sensitive to the developmental and methodological issues inherent in growth measurement are needed to confirm our findings.
|
ACKNOWLEDGMENTS |
|---|
Dr Biederman receives research support, is a speaker, or is on the advisory board for the following: Cell Tech and Shire Laboratories, Inc, Eli Lilly & Co, Wyeth Ayerst, Pfizer Pharmaceutical, Cephalon Pharmaceutical, Janssen Pharmaceutical, Noven Pharmaceutical, GlaxoSmithKline, Alza/McNeil Pharmaceuticals, Stanley Foundation, National Institute of Mental Health, National Institute of Child Health and Development, and National Institute on Drug Abuse.
|
FOOTNOTES |
|---|
Reprint requests to (J.B.) Pediatric Psychopharmacology Unit (ACC 725), Massachusetts General Hospital, Fruit St, Boston, MA 02114. E-mail: jbiederman{at}partners.org
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  T. Hanc and J. Cieslik Growth in Stimulant-naive Children With Attention-Deficit/Hyperactivity Disorder Using Cross-sectional and Longitudinal Approaches Pediatrics, April 1, 2008; 121(4): e967 - e974. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. J. Spencer, J. Biederman, J. H. Newcorn, C. J. Kratochvil, D. Ruff, and D. Michelson Absence of Unfavorable Effect of Atomoxetine on Growth in Pediatric Patients Is Still Not Demonstrated: In Reply Pediatrics, February 1, 2006; 117(2): 588 - 588. [Full Text] [PDF]
|
|
|
|
 |
|
 A Poulton Growth on stimulant medication; clarifying the confusion: a review Arch. Dis. Child., August 1, 2005; 90(8): 801 - 806. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V A Harpin The effect of ADHD on the life of an individual, their family, and community from preschool to adult life Arch. Dis. Child., February 1, 2005; 90(suppl_1): i2 - i7. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. K. Mulhern, R. B. Khan, S. Kaplan, S. Helton, R. Christensen, M. Bonner, R. Brown, X. Xiong, S. Wu, S. Gururangan, et al. Short-Term Efficacy of Methylphenidate: A Randomized, Double-Blind, Placebo-Controlled Trial Among Survivors of Childhood Cancer J. Clin. Oncol., December 1, 2004; 22(23): 4795 - 4803. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Does ADHD Treatment Suppress Growth in Girls? Journal Watch Psychiatry, October 8, 2003; 2003(1008): 10 - 10. [Full Text]
|
|
|
|
 |
|
 Does ADHD Treatment Suppress Growth in Girls? Journal Watch Pediatrics and Adolescent Medicine, July 28, 2003; 2003(728): 2 - 2. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2003 American Academy of Pediatrics. All rights reserved. |
||
|
||
